Proinflammatory versus anti-inflammatory response in sepsis patients: looking at the cytokines

Despite improvements in supportive care, mortality rates in sepsis remain substantially high. Sepsis exhibits phases of enhanced inflammation, alternating with immune suppression with a resultant variant time point of mortality; yet no human study has correlated levels of cytokines to the timeline of mortality. Our study attempts to analyze the association of levels of proinflammatory and anti-inflammatory cytokines in sepsis with the timeline of death in terms of early ( 5 days) mortality, and day of death. We also assessed correlation of these cytokines with length of stay.

Introduction: Among methods for preventing pneumonia and possibly also bacteremia in ICU patients, selective digestive decontamination (SDD; topical with or without protocolized parenteral antibiotic) appears most effective within randomized concurrent controlled trials (RCCTs) [1]. However, whether parenteral antibiotic is required, and whether SDD actually increases pneumonia incidences in SDD RCCTs versus the broader ICU pneumonia evidence base, remain unresolved [2,3]. The purpose of this analysis is to test for counterfactual and contextual effects of the topical and parenteral SDD components on the bacteremia incidence versus the broader evidence base related to the patient group at risk of VAP. Methods: Bacteremia incidence proportion data were extracted from component (control and intervention) groups from studies investigating antibiotic (SDD) or nonantibiotic methods of VAP prevention. Both the counterfactual and the contextual effects of SDD factorized as topical or protocolized parenteral exposures were estimated using random-effects meta-analysis of study and group level data. Studies without any prevention methods under study constituted the reference category (benchmark groups). Results: As a counterfactual within RCCTs, SDD when given as combined topical and parenteral antibiotic appears to halve the bacteremia incidence (odds ratio (OR) 0.59; 0.48 to 0.73; n = 18 studies). As a contextual however, the mean bacteremia incidence among 27 control groups (17.1%; 13.1 to 22.1%) and 12 intervention groups receiving topical antibiotic alone (16.2%; 9.1 to 27.3%) from SDD RCCTs is double that of 36 benchmark groups (8.3; 7.0 to 10.8%) and 19 control groups from studies of nonantibiotic methods (7.7%; 5.2 to 11.1). The upward dispersion in bacteremia incidence among component groups from SDD RCCTs away from this benchmark is striking with all but two of the 27 control groups and all but two of 12 SDD intervention groups that did not receive PPAP being above this benchmark. Conclusion: The major contextual hazard of SDD toward bacteremia among ICU patients is inapparent within individual studies. The apparent protection in SDD RCCTs is spurious as the SDD counterfactual is conflated by the strong contextual effect with partial mitigation by SDD via intranasal route, showed significant (P ≤ 0.05) decrease in neutrophil influx into the lungs. A significant (P ≤ 0.05) decrease in the production of proinflammatory cytokines (that is, TNFα and IL-1α) and other biochemical mediators of acute inflammation (that is, MDA, MPO, and NO) was also observed in this group. But the augmentin treatment alone did not decrease these proinflammatory mediators significantly (P ≥0.05) as compared to the control group. Conclusion: We therefore conclude that thalidomide ameliorates lung inflammation induced by K. pneumoniae B5055 without significantly (P ≤ 0.05) decreasing the bacterial load in the lung tissue whereas augmentin takes care of bacterial proliferation. Hence, it can be used as an adjunct therapy along with antibiotics as an anti-inflammatory or an immunomodulatory agent in case of acute lung infection or pneumonia. Introduction: Purulent complications in patients with type 2 diabetes are usually severe, often complicated by sepsis and require emergency surgery. Noncardiac surgery is associated with a 7 to 11% complication rate and mortality of 0.8 to 1.5% [1], up to 42% are cardiac reasons [2]. After surgery, 2% of patients suffer major cardiac complications [3], and 8% show evidence of significant myocardial injury [2]. The aim of this study was to identify the impact of purulent complications and sepsis on cardiovascular system in patients with type 2 diabetes. Methods: We analyzed 112 consecutive patients (54 men and 58 women) aged 57.2 ± 8.4 years with purulent-necrotic complications (gangrene, phlegmon, and abscess) of type 2 diabetes and sepsis in 2013. We compared laboratory and instrumental data (blood tests, ECG, echocardiography and others), which were previously obtained in the same patients receiving inpatient treatment before sepsis (2011 to 2012). Results: Gangrene of lower extremities in 59 (52.7%) prevailed among purulent complications. After the development of sepsis we detected in all patients significantly increased heart rate, respiratory rate per minute, leukocytosis, anemia, worse glucose metabolism and renal function (Table 1). Congestive heart failure became more severe. This was confirmed by decrease of left ventricle ejection fraction (55.2 ± 5.1% before sepsis vs. 49.3 ± 4.1% after) and increase brain natriuretic peptide (291.4 ± 34.5 ng/ml vs. 395.2 ± 28.1 ng/ml, P < 0.001). Prior sepsis in 66 (58.9%) of patients with arterial hypertension was observed, after in 88 (78.6%). After admission to the centre, patients had no signs of septic shock. In 13 (11.6%) patients, the perioperative period was complicated by acute myocardial infarction, which was accompanied by a fall in blood pressure. We detected an increase of the functional class of stable angina, congestive heart failure, 4.2 times increased incidence of unstable angina, 2.6 times ventricular and four times supraventricular extra systole after septic complications (Table 2). Conclusion: After the development of purulent complications and sepsis in patients with type 2 diabetes, we observed increased incidence of arterial hypertension, arrhythmias, worsened severity of coronary artery disease and congestive heart failure. Perioperative risk of acute myocardial infarction amounted to 11.6%. Introduction: Lung abscesses and gangrene are the most severe clinical manifestation and outcome among acute purulent destructive pulmonary disease (APDPD). Mortality ranges from 10 to 35%, and in the presence of diabetes increases up to 30 to 90% [1]. The main reason for this is the generalization of infection (sepsis), leading to the development of multiple organ failure [2,3]. The aim of this study was to identify the severity of sepsis in patients with APDPD depending on the presence of type 2 diabetes, and the impact on the forecast. Methods: During the period 2012 to 2013, we examined 408 patients aged 48.5 ± 12.5 years (258 men/150 women) who underwent surgical treatment for APDPD. The patients were divided into two groups: 144 patients with type 2 diabetes, and controls (n = 246). We carried out computed tomography, ECG, echocardiography, laboratory biochemical testing, and bacteriological analysis of pathologic material and blood samples. Results: Patients with type 2 diabetes had much more complications and cases of severe sepsis and septic shock (Table 1). Bacteriological analysis of the pathologic material showed Gram-positive bacteria in 35 to 45%, anaerobic association in 55 to 65%, pathological fungi in 50 to 60%. The patients with type 2 diabetes had much more time from the onset of the first symptoms of lung disease prior to admission (12.5 ± 3.5 vs. 7.5 ± 2.5 days, P = 0.002), and the duration of inpatient treatment was significantly longer (13.8 ± 5.5 vs. 7.1 ± 3.4 days, P = 0.001). Only 53 (36.8%) of patients with type 2 diabetes and 68 (29.5%) without it had bacteriological positive blood culture. The analysis of the distribution of pathogens in groups is presented in Figure 1. Patients with diabetes had more Candida spp.  Figure 1). Figures 2, Figure 3 and Figure 4 present the X-ray dynamics of a 42-year-old man with lung abscess. Clinical recovery in patients with type 2 diabetes was significantly worthy compared to controls (45 (31.2%) vs. 153 (57.9%)), mortality rate 48 (33.3%) versus 39 (14.7%), respectively. Conclusion: In patients with acute purulent destructive pulmonary disease and type 2 diabetes, severe sepsis and septic shock more often prevailed, inpatient mortality rate was 2.27 times higher, compared to patients with normal glucose metabolism.  To analyze the risk factors, we include 31 patients with lung abscess and sepsis complicated by mediastinitis in the first group, and 187 patients without mediastenitis in the second group. Groups were similar in age (46.1 ± 8.2 years vs. 45.8 ± 13.2 years). A total 77.4% of patients with mediastinitis were women who suffered from type 2 diabetes (HbA1c = 9.7 ± 1.4%), congestive heart failure and anemia. Significant differences in the groups according to the data of laboratory and instrumental studies are presented in Table 1.
Conclusion: In total, 14.2% of patients presented with lung abscess and sepsis complicated by mediastinitis, more commonly in women with diabetes mellitus, obesity, anemia and reduced ejection fraction of the left ventricle. Introduction: The minimal biological active structure of endotoxins (lipopolysaccharides (LPS)) is Re-LPS (KDO2-lipid A), which consist of lipid A and two (or three) molecules of 3-deoxy-D-manno-2-octulosonic acid (KDO) [1,2]. Biological activity of endotoxins is defined in general by the number and distribution of acyl residues on the lipid A backbone [3].
Recently it has been reported that KDO-treated RAW 264.7 cells exhibited a gene expression pattern similar to that in LPS-treated cells. These authors revealed that free KDO participated in crosstalk between Toll-like receptors (TLR) and G protein-coupled receptors and so that regulated activators and repressors of immune signaling [4]. LPS-dependent TLR4-triggered activation of target cells leads to specific changes in the levels of surface receptors and induces synthesis of proinflammatory cytokines [5]. However, the dependence of these processes on the structural composition of LPS is not well understood. To extend our knowledge in this field, the effects of free KDO as well as KDO as covalently linked to lipid A constituent of Re-LPS on expression of TLR4, CD11b and CD14 receptors and TNFα synthesis in whole human blood have been investigated. Methods: Human blood was incubated with Re-LPS from Escherichia coli JM103 or Salmonella enterica sv Typhimurium SL1181 (100 ng/ml) or with lipids A from E. coli F583 or S. enterica sv Minnesota R595 (80 ng/ml) or with ammonium salt of KDO (20 ng/ml) at 37°C in 5% CO 2 -humidified atmosphere for 2 or 6 hours to determine receptor expression or TNFα release, respectively. Receptor expression was monitored by EPICS XL-MCL flow cytometer using Alexa Fluor 488 anti-TLR4 (HTA125), anti-CD11b (ICRF44) and anti-CD14 (HCD14) antibodies. Human TNF-α ELISA Kit II was exploited to TNFα determination.
Results: Re-LPS E. coli or Re-LPS S. enterica differentially affected receptor expression in comparison to their respective lipids A. Free KDO in the equimolar concentration as it exists in KDO2-lipid A (Re-LPS) did not influence the level of CD14 but downregulated the expression of TLR4 and CD11b ( Figure 1). Tenfold increased KDO concentration did not affect further the receptor expression. The addition of KDO2 to lipid A E. coli -that is, applying KDO as covalently linked constituent of Re-LPS -led to upregulation of CD14 and TLR4 but downregulated CD11b expression. The expression of TLR4 was most pronounced upregulated by Re-LPS S. enterica but in the case of CD14 and CD11b this Re-LPS had an opposite effect in comparison to E. coli endotoxins (table in Figure 1). Lipid A S. enterica was a less potent TNFα inductor than that from E. coli ( Figure 2). This may be explained by the differences in lipid A composition determining lipid A affinity to target receptor(s). LPS E. coli, as had been shown early, caused MyD88-dependent fast NF-B degradation (rapid TNFα response) whereas LPS S. enterica induced MyD88-independent signaling (delayed TNFα response) [5]. In our study, free KDO did not stimulate TNFα release. KDO2 as a constituent of Re-LPS S. enterica increased significantly the TNFαinducing activity of lipid A S. enterica but this effect was not so distinguished between Re-LPS E. coli and lipid A E. coli ( Figure 2). Conclusion: Free KDO in the used concentration was inactive in regulation of TLR4, CD11b and CD14 expression and did not induce TNFα release but its impact in biological activity was detected when KDO was applied as constituent of Re-LPS. This may be explained by the effect of KDO on the spatial conformation of Re-LPS. Introduction: Inflammation is a host defense reaction against pathogenic infection. In this process, inflammatory cytokines contribute to combat against infection, but excess cytokines will lead to tissue damage. Nonsteroidal anti-inflammatory drugs and corticosteroids are commonly used for regulating these inflammatory mediators and treatment of inflammatory disorders. But these drugs are not sufficient for clinical practice due to their adverse effects, so new anti-inflammatory drugs are still needed. Evodiamine (EVO), an important alkaloidal component extracted from the fruit of Evodiae fructus, possesses the property of analgesia, antiemesis, and vascular dilatation. Its anti-inflammatory effect and underlying mechanism were investigated using a zymosan-induced inflammatory model. Methods: In vitro, RAW264.7 cells and primary peritoneal macrophages were treated with different doses of EVO (25, 50 or 100 μM) for 1 hour prior to incubation with zymosan (100 μg/ml), and the effects of EVO on protein and mRNA levels of proinflammatory cytokines were determined by ELISA and qRT-PCR, respectively. In vivo, peritonitis was induced in C57BL/6 mice by zymosan (500 mg/kg) injection, and the effects of EVO (10 mg/kg) on plasma cytokine levels and organ injury were evaluated. Activation of the NF-B signal pathway was investigated by ELISA-based Trans-AM transcription factor NF-B p65 kit, immunocytochemistry and western blotting. Results: EVO effectively suppressed the expression of IL-1β, IL-6 as well as TNFα in both protein and mRNA level in vitro. It can also attenuate zymosan-induced DNA-binding activity of NF-B, which was achieved through the inhibitory effects on the phosphorylation of inhibitory kappa B α (IBα) and p65 nuclear translocation, but there was little association with mitogen-activated protein kinase activation. In vivo, treatment with EVO could ameliorate inflammatory cell infiltration and vascular ectasia induced by zymosan in both lung and intestine tissues. EVO can markedly decrease the level of TNFα and IL-6 in plasma, and effectively downregulate expression of IL-6, TNFα and myeloperoxidase in both lung and intestine. Moreover, cell apoptosis in organs was also attenuated by treatment of EVO. The underlying mechanism that a decrease in the phosphorylation of IBα and the subsequent transcription activity of NF-B was also confirmed. Conclusion: Taken together, our data demonstrate that EVO displays antiinflammatory actions in vitro and in vivo by suppressing the phosphorylation of IBα and inactivating NF-B, which suggests that EVO is a potential therapeutic agent against inflammatory disorder.

P12
Imaging in severe sepsis and septic shock: is early radiological identification of occult sources of infection needed?. Introduction: The importance of imaging in establishing the focus of infection is recognised in current guidelines for the management of severe sepsis [1], with decisions regarding timing and modality of imaging left to the physicians' clinical judgement. In the emergency department (ED), clinical assessment combined with bedside investigations of chest X-ray (CXR) and urine dip can be used to confirm the two most common sources [2]. However, they may fail to identify occult sources of infection, such as intraabdominal collections and abscesses, the treatment of which may require alteration of empirical treatment or be refractory to antibiotic therapy alone. Further imaging is necessary to confirm the focus so that optimal treatment can be achieved. Methods: The study cohort was composed of 50 consecutive patients who met the criteria for severe sepsis [1] attending the ED in 2013. Electronic and paper patient records and radiology results were analysed. All radiological studies done in the first 72 hours following attendance were included in the study.
Results: CXR was performed as the initial investigation in 49 of the 50 patients (98%). The median time from arrival at the ED to initial imaging was 1 hour:00 minutes (range 0 hours:04 minutes to 4 hours:25 minutes). Initial investigations in the ED of CXR and urine dip identified a septic focus in 30 of 50 patients (60%). Fourteen of the remaining 20 went on to have one or more further imaging studies. Figure 1  Introduction: Despite improvements in supportive care, mortality rates in sepsis remain substantially high. Sepsis exhibits phases of enhanced inflammation, alternating with immune suppression with a resultant variant time point of mortality; yet no human study has correlated levels of cytokines to the timeline of mortality. Our study attempts to analyze the association of levels of proinflammatory and anti-inflammatory cytokines in sepsis with the timeline of death in terms of early (<5 days) versus late (>5 days) mortality, and day of death. We also assessed correlation of these cytokines with length of stay. Methods: The study protocol was approved by Institutional Ethics Committee. Subjects were 74 consecutive patients with communityacquired severe sepsis/septic shock admitted to the ICU of a tertiary care superspeciality hospital. Blood samples drawn on days 1, 3 and 7 of admission were analysed for proinflammatory cytokine (TNFα) by chemiluminescent immunometric assay and anti-inflammatory cytokine (IL-10) by ELISA. Subjects were segregated on basis of: ratio of proinflammatory and anti-inflammatory mediators on day 1 of admission into patients with predominant proinflammatory or predominant anti-inflammatory response. Survival and time point of mortality into survivor, early mortality and late mortality groups. Statistical analyses were performed using SPSS version 17. Results: There were 37 patients each in predominant proinflammatory and predominant anti-inflammatory groups. The number of deaths was 11 and 17 respectively in these groups. However, there was significantly higher early mortality in the proinflammatory group as compared to the antiinflammatory group (7 vs. 1, P = 0.0247). On the contrary, late deaths were significantly higher in the anti-inflammatory group as compared to the proinflammatory (16 vs. 4 P = 0.0017). The ratio of proinflammatory/antiinflammatory cytokines (TNF/IL-10) on day 1 was significantly lower in patients of late death (n = 20) as compared to patients of early death (n = 8) and survivors (n = 46) as shown in Table 1. Further, the median day of death  Results: We identified a leukocyte gene expression signature that positively correlated with S. aureus disease severity [1]. This severity signature was enriched for genes associated with neutrophils but was not solely explained by increased percentage of neutrophils. This set of genes was also associated with severity in sepsis, with higher expression in patients with septic shock compared with sepsis [2] and in nonsurvivors compared with survivors of septic shock [3]. Our in vitro studies revealed that the severity signature may reflect an increase in circulating immature neutrophils or band cells which has been previously reported in the context of bacterial stimuli and sepsis [4,5]. This line of evidence is consistent with a recent report by Guerin and colleagues that demonstrated that quantification of immature neutrophils by flow cytometry was prognostic for sepsis mortality [6]. Introduction: Autophagy could be induced under stress conditions, including starvation, infection, and hypoxia. The microRNA (miRNA) network may be critical in the regulation of autophagy. Upregulation of autophagy may be a protective response for cell survival in ischemic kidney injury. The aim of this study was to evaluate whether miRNA regulates autophagy in ischemic kidney injury and renal proximal tubular cells under hypoxic conditions. Methods: Ischemic kidney injury was performed by clamping bilateral renal pedicles for 60 minutes in male mice. Human kidney proximal tubular (HK2) cells are incubated in a hypoxic chamber with 0.3% O 2 . Bioinformatics analyses were used to select the candidate miRNA. Gainof-function and loss-of-function methods were employed to evaluate the effects of miRNA on autophagy. Chromatin immunoprecipitation analyses and promoter luciferase reporter assays were used to evaluate the interaction of transcriptional factors with miRNA. Results: Increase of LC3 and ATG16L1, autophagy-related proteins, and down expression of miR-20a-5p were detected in kidneys after ischemic injury and in HK2 cells under hypoxic conditions. The 3'-untranslated region luciferase reporter assays indicated that miR-20a-5p targeted ATG16L1 messenger RNA.
Overexpression of miR-20a-5p reduced the expression of LC3-II and ATG16L1 in HK2 cells under hypoxic conditions, whereas antagomiR-20a reversed the inhibition. Using RNAi against hypoxia-inducible factor-1α (HIF-1α) in HK2 cells, we confirmed the inhibitory binding of HIF-1α to miR-20a-5p. Conclusion: The signaling axis of HIF-1α, miR-20a-5p, and ATG16L1 in autophagic process might be a critical adapting mechanism for ischemic kidney injury. Further studies need to be conducted to validate the sensitivity and specificity of the SSST; changes will be recommended in an effort to improve sensitivity.
Introduction: Compliance with the Surviving Sepsis Campaign 2012 (SSC) bundle in the emergency department (ED) is a key point to improve outcome of severe sepsis and septic shock [1,2]. Before and after education of ED staff, we registered compliance and timing of lactate dosing, blood culture sampling, empiric antibiotic therapy (ATB) and fluid resuscitation, the 3-hour (H3) bundle. Survival and compliance according to the initial pathway of care were also studied. Methods: A monocentric study before and after education of ED staff about SSC bundles (courses, posters, pocket guides). We looked at compliance of the H3 bundle items in a retrospective and a prospective cohort, timing of realisation, day 28 survival, overall severity (SAPS2, SOFA and RISCC scores), impact of prehospital medical management, and initial pathway of care. Statistical analysis was performed with Fisher exact test and Mann-Whitney test. Multivariate analysis of factors associated with survival was made through logistic regression. Results: Eighty-nine patients were included in the prospective cohort, 65 in the retrospective cohort. Patterns of patients in the retrospective and prospective cohort were respectively: sex ratio M/F 29/39 and 39/47 (NS);  , and associated with a low SAPS2 score in multivariate analysis. Admission through a prehospital medical team was associated with a stronger H3 ATB compliance before intervention (P = 0.032). Within the ED, initial orientation to the acute care unit was associated with a better H3 ATB compliance compared to standard care before and after staff education (P = 0.001; P = 0.003), and with better overall compliance (P = 0.004; P = 0.026). Conclusion: Compliance with the SSC H3 bundle was increased but still needs to be improved. There is an impact of the initial pathway of care on compliance of the bundle, and on timing of ATB injection. Differences in healthworker/patient ratio in the units of care could explain these disparities [3]. Improvement could be obtained through optimizing early screening, correct initial guidance, or with dedicated teams.
compliance with the resuscitation bundle in the management of severe sepsis and septic shock. Crit Care 2013, 17:R224.

P18
Benefit of achieving lactate clearance versus central venous oxygen saturation target as microcirculation end point resuscitation in severe sepsis and septic shock. Introduction: In severe sepsis and septic shock patients, lactate clearance >10% and central venous oxygen saturation (ScvO 2 ) >70% are accepted parameters of tissue oxygenation adequacy. There is controversy of which parameters better associate with early mortality, and thus should be implemented as the microcirculation end point resuscitation [1][2][3]. This study was aimed to address the association of achieving either one or two targets of microcirculatory end point resuscitation and early mortality in severe sepsis and septic shock patients.
Methods: A retrospective cohort study was conducted in severe sepsis and septic shock patients (aged 18 years and older) hospitalized in the ICU, Cipto Mangunkusumo Hospital, Indonesia. Patients' early outcomes were observed during first 120 hours of hospitalization. Cox's regression analysis was used to analyse risk of early mortality in subject groups achieving lactate clearance target only, ScvO 2 target only, both targets, and not achieving any target in 6 hours after onset of resuscitation.
Results: Subjects consisted of 268 patients. Early mortality developed in 70 subjects. Fifty-four subjects achieved lactate clearance target only, 16 achieved ScvO 2 target only, 138 achieved both targets, 60 did not achieve any target. Subjects who achieved both targets had a significant lowest early mortality risk (P = 0.104 compared with subjects achieved lactate clearance target only and P = 0.000 compared with remaining subject groups) ( Figure 1). In subgroup analysis of subjects who achieved lactate clearance or ScvO 2 target only, failure to achieve lactate clearance target associated with higher early mortality risk (hazard ratio 5      Introduction: Sepsis is one of the leading causes of death around the world. The failure of clinical trials to treat sepsis demonstrates that the molecular mechanisms are multiple and still insufficiently understood. The objective is to clarify the long disputed hierarchical contribution of several central inflammatory mediators, namely IL-1β, IL-18, CASP7, CASP1 and CASP11, in septic shock, and to explore their therapeutic potential. Methods: LPS-induced and TNF-induced lethal shock, as well as cecal ligation and puncture (CLP), were performed in genetically or pharmacologically targeted mice. Body temperature and survival were monitored closely, and plasma was analyzed for several markers of cellular disintegration and inflammation.
Results: Interestingly, deficiency of both IL-1β and IL-18 additively prevented LPS-induced mortality. The detrimental role of IL-1β and IL-18 was confirmed in mice subjected to a lethal dose of TNF, or to a lethal CLP procedure. Although their upstream activator, CASP1, and its amplifier, CASP11, are considered potential therapeutic targets because of their crucial involvement in endotoxin-induced toxicity, CASP11 or CASP1/11 deficient mice were not, or hardly, protected against a lethal TNF or CLP challenge. In line with our results obtained in genetically deficient mice, only the combined neutralization of IL-1 and IL-18, using the IL-1 receptor antagonist Anakinra and anti-IL-18 antibodies, conferred complete protection against endotoxin-induced lethality. Conclusion: Our data point towards the therapeutic potential of neutralizing IL-1 and IL-18 simultaneously in sepsis, rather than inhibiting the upstream inflammatory caspases. Introduction: Severe sepsis and septic shock (SS/SS) have a high mortality. Therapeutic guidelines can improve mortality, but early recognition and timely implementation of these requires a proactive attitude that can be electronically supported.
Methods: From May 2013 our hospital implemented a Sepsis Code (SC) based on an early detection electronic rule developed by our multiprofessional sepsis team: clinicians and IT engineers (EMR Cerner Millennium platform) and a standardized order set plus systematic follow-up by our sepsis team. We performed a before-after study to assess the impact over mortality of this strategy. Time-series analysis of sepsis admissions and mortality from January 2011 to December 2013, before and after SC implementation. (Analysis by STATA.) All urgent admissions recorded in the minimum basic data set in patients over 14 years from 1 January 2011 to 31 December 2013 were included. Inclusion criteria: patients with ICD-9 sepsisassociated codes in the principal diagnosis or patients with infectionassociated codes in the principal diagnosis together with sepsis-associated codes in secondary diagnosis. Medical records were manually reviewed by clinicians to confirm SS/SS diagnosis. Temporal series analysis (Poisson regression). First analysis: sepsis admissions in relation to total urgent admissions. Second analysis: deaths due to SS/SS related to admissions in this group. In both cases we compared results before SC activation (28 months) and after that (first 2 transitional months and 6 consolidated months). The multivariate adjustment in both analyses included year, month of the year, and months with activated rule. Graphic analysis estimated predictions for the last 8 months based on the previous 28 months, comparing both observed and predicted sepsis and deaths.
Results: A total of 24,118 urgent admissions were included, 5,657 in the postalert period. Mean monthly admissions: 652 (SD 47) (570 to 740). In total, 408 and 178 SS/SS were identified in the prealert and postalert period, respectively. After SC implementation we observed no significant changes in sepsis admission risk but a clear downward trend in sepsis mortality: in the first 2 transitional months we did not observe major changes, while in the last 6 months the risk of death does fall 36% reaching statistical significance (IRR 0.64 (95% CI 0.43 to 0.97, P = 0.036)) ( Table 1 Figures 1 and Figure 2). Both antibiotic door-to-needle time and adequacy significantly improved in sepsis cases where the alert was triggered.
Conclusion: Implementation of a SC triggered by an electronic detection alert, compared to the prealert period, decreased mortality risk by 36% (IRR 0.64 (95% CI 0.43 to 0.97, P = 0.036)) with the rule fully implemented.

P22
Assessing the value of a real-time electronic screening algorithm for early detection of severe sepsis in the emergency department. Introduction: In severe sepsis/septic shock (SS/SS), early recognition and timely implementation of treatment is critical for survival, and this could be electronically supported. We assess the value of an electronic automatic algorithm based on EMR data as a screening tool for early detection of sepsis.
Methods: Our multiprofessional sepsis team (clinicians and IT engineers) developed an electronic algorithm using data from our EMR (Cerner Millennium platform) aimed at the automatic, real-time recognition of two or more systemic inflammatory response syndrome components + one or more organ failure parameter (according to sepsis definition) in every patient attended in the ER. The firing of this sepsis rule issues an alert to the responsible clinician to confirm an infectious etiology and opens an electronic standardized order set according to sepsis bundles. The alert database (from its start in May 2013 to December 2013) was cross-matched with the minimum basic data set for urgent admissions (>14 years) during this same period. We selected, based on an ad hoc syntaxes, those admissions due to sepsis. Medical records were manually reviewed for confirmation of SS/SS. We assessed sensibility, specificity, negative and positive predictive value of the electronic rule, considering the confirmed clinical diagnosis at discharge as the gold standard.
Results: In total, 37,323 patients were seen in the ER, 5,657 emergency admissions took place and 178 were due to SS/SS. Alert fired in 1,190 (3.2%) total emergencies and in 754 emergency admissions (13.3%). Data analysis after alert implementation identifies a global sensitivity of 80%, which improved after the first 2 months of transition. In the last 6 months (consolidated period) it was between 85 and 90%. Global specificity 89%, NPV of 99% and PPV of 19% for a global prevalence at admission of 3.2 cases/100 ( Introduction: Although there have been many clinical trials for treatment of sepsis, no drugs are available at present. Sepsis is thought to be complex disorders; activation/lesion of vascular endothelial cells, accelerated coagulation and platelet aggregation, leukocyte activation or paralysis, and hypercytokinemia. To treat the pathological status of sepsis, it must be necessary to control these plural disorders simultaneously or sequentially. In the present study, we identified and characterized a plasma protein histidine-rich glycoprotein (HRG) as a factor that decreases dramatically in septic condition and maintains neutrophil's shape and functional quiescence. We clarified the involvement of HRG in septic pathogenesis and propose a novel therapy for sepsis based on that.
Methods: Sepsis was induced in mice by cecal ligation and puncture. The mice were treated with HRG purified from human plasma after confirming the marked decrease in plasma HRG. We evaluated the beneficial effects of HRG administration on survival rate, lung inflammation, and the state of circulating neutrophils using in vivo imaging. Purified neutrophils from human blood were treated with HRG and analyzed with respect to neutrophil shape, adhesiveness to vascular endothelial cells, passage through microcapillaries, production of reactive oxygen species, and cytoskeleton rearrangement with relevant signal transduction.
Results: Supplementary treatment of septic mice with exogenous HRG for the decrease in plasma HRG improved the survival of mice remarkably. HRG treatment reduced the number of neutrophils in the lung, on which platelet aggregation and fibrin deposit were observed. HRG also inhibited the expression of mRNAs of IL-6, TNFα, iNOS, and PAI-1 in the lung. In contrast, knockdown of HRG by siRNA exacerbated lethality. Purified human HRG reversibly induced morphological changes in human neutrophils in vitro; induction of spherical shape with reduced microvilli and adhesiveness to vascular endothelial cells. HRG maintained the passage of neutrophils through microcapillaries and abolished the production of reactive oxygen species whereas HRG had no effect on the expression of adhesion molecules including CD11b and CD62L. Conclusion: We show that plasma protein HRG is a crucial factor that keeps the circulating neutrophils quiescent and prevents unnecessary activation in bloodstream using a cecal ligation and puncture model in Introduction: Intravenous immunoglobulin (IVIG) has been used as adjuvant therapy for severe sepsis patients expecting the anti-inflammatory effect. However, the effects of IVIG have been controversial. The majority of ill patients with SIRS had coagulation abnormalities. In addition, inflammation and coagulation play pivotal roles in the pathogenesis of sepsis. Moreover, the evidence of extensive cross-talk between these two systems has been increasing. The aim of this study is to investigate the effects of IVIG treatment for inflammation and hemostatic abnormality in sepsis patients.
Methods: This prospective single-center observational study was conducted in our ICU between January and July 2013. We enrolled 41 patients (≥18 years, admitted to the ICU diagnosed for sepsis, and more than 7 days of ICU stay) and divided them into two groups: IVIGtreated group (IVIG group) and non-IVIG-treated group (non-IVIG group). After that, we compared inflammatory molecule markers (WBC, CRP, procalcitonin (PCT), and interleukin-6 (IL-6)), coagulation/fibrinolysis markers (platelet counts, PT-INR, APTT, D-dimer, TAT, PIC, soluble fibrin (SF), and plasminogen activator inhibitor-1 (PAI-1)), and Japanese Association for Acute Medicine Disseminated Intravascular Coagulation (JAAM DIC) score and positive rate at the admission period (day 1) and days 4 to 7 between two groups. Moreover, the 28-day mortality rate was investigated.
Results: Nineteen patients were treated with IVIG (IVIG group) and 22 patients were without (non-IVIG group). In the IVIG group, PCT and IL-6 were significantly higher than that in the non-IVIG group at day 1 (P < 0.01, and P < 0.01). In this group, CRP, PCT, and IL-6 were significantly decreased at days 4 to 7 rather than that at day 1 (P < 0.01, P < 0.01, P < 0.01, respectively). Moreover, the JAAM DIC score was decreased at days 4 to 7 than that at day 1 significantly (P < 0.05) in this group. We also confirmed that PT-INR, APTT, TAT, SF, and PAI-1 were significantly improved between day 1 and days 4 to 7. On the other hand, in the non-IVIG group there was significantly decreased IL-6 and TAT only, but not CRP, PCT, PT-INR, APTT, SF, PAI-1 and JAAM DIC score between day 1 and days 4 to 7. For the 28-day mortality rate, the IVIG group was lower than that of the non-IVIG group (IVIG group, 5.3%; non-IVIG group, 18.2%). Conclusion: Our study demonstrated that IVIG treatment significantly improved the inflammatory and hemostatic abnormalities in sepsis patients.  Introduction: The optimal evaluation of fluid resuscitation in patients with acute respiratory distress syndrome (ARDS) remains to be clearly elucidated. The purpose of this study was to clarify the influence of fluid overload in patients with ARDS. Methods: Two hundred and seven ARDS patients admitted to the ICUs of 23 tertiary referral hospitals in Japan were enrolled in this study. These patients were divided into survivor (survival group, n = 137) or nonsurvivor (nonsurvival group, n = 70) groups according to the 28-day mortality. All patients received mechanical ventilation and also underwent transpulmonary thermodilution monitoring. The data for analysis were collected for three consecutive days from time of admission. Results: On the second hospital day, the extravascular lung water index of the nonsurvival group was significantly higher than that of the survival group (18.6 ± 9.4 ml/kg vs. 15.4 ± 6.3 ml/kg, P = 0.03). Moreover, regarding the first 3-day cumulative fluid balance, the nonsurvival group was significantly higher than survival group (5.1 ± 4.3 l vs. 3.5 ± 0.4 l, P = 0.015). We suspected that these results might be related to the cardiovascular and/ or renal function. We therefore excluded any patients with a score of three or more regarding the cardiovascular and/or renal criteria about the Sequential Organ Failure Assessment score. Thereafter, we confirmed the results to be similar for the first 3-day cumulative fluid balance between the two groups (3.8 ± 1.6 l vs. 2.2 ± 4.0 l, P = 0.0339). A stepwise logistic regression analysis identified the 3-day cumulative fluid balance to be an independent risk factor for the 28-day mortality (adjusted odds ratio: Introduction: Systemic infection of the human host can arise from several pathogenic bacteria as well as from fungal species, and it is of high clinical relevance to trace back the nature of infection from the host response. Amongst these systemic infections are sepsis inducers Candida albicans and Mycobacterium tuberculosis, responsible for a great amount of worldwide deaths [1]. Sepsis, however, does not necessarily originate from infection. Traumas and non-infection-based injuries can also trigger an unbound inflammatory response from the human host most commonly known as systemic inflammatory response syndrome. Both cases, infection and noninfection, nevertheless display similar clinical symptoms with significantly better recovery times of patients included in the latter. Despite the clinical similarities, studies have suggested that distinct infection-induced host responses from different pathogens occur, namely at the signalling pathway level [2]. Methods: Studies have been performed focused on common gene expression profiles between several pathogens [3]. However, understanding the difference in immune responses between these two pathogens, but not exclusively, might lead to better diagnostic tools and treatment decisionmaking. Relying on systems biology concepts and bioinformatics tools, we use gene expression data to distinguish C. albicans and M. tuberculosis infections in the human host; for example, cellular regulation and communication between host immune cells and pathogens [4,5]. Results: We have, in a first analysis, focused on the cytokine-cytokine signalling pathway due to its role in inflammation response towards infections. Genes belonging to the IL-2 cytokine family are only expressed when facing infection by C. albicans in comparison to M. tuberculosis suggesting a tendency towards B-cell proliferation and production of antibodies. However, during infection caused by M. tuberculosis, no significant changes in gene expression occur in this pathway that indicates a specific immune response for this pathogen. Further analysis of additional signalling pathways might highlight other human infection-specific immune responses in regards to the pathogens considered. Conclusion: Next we will focus on developing a mathematical model capable of simulating such immune responses and possibly identifying genes and pathways which might clarify how these inflammation responses can be targeted, countered and moderated [6]. Introduction: Systemic salmonella infection is a frequent cause of Gramnegative sepsis. Bacterial lipopolysaccharide in blood triggers immune response by monocytes, which results in overwhelming production of proinflammatory cytokines and pathology in peripheral organs such as the liver, heart and lungs. The mortality rate due to sepsis remains high even after chemotherapeutic clearance of pathogens, due to sustained production of inflammatory mediators. Therefore, anti-inflammatory therapy as an adjunct to antibiotics could reduce the mortality from sepsis. Methods: To date, several studies have evaluated the role of vasoactive intestinal peptide (VIP) as a therapeutic agent in sepsis both in vivo and in vitro since it possesses several desirable biological properties. The peptide, acting via VIP cell receptors, mediates effects by altering production and secretion of inflammatory mediators such as cytokines. VIP has been shown to decrease production of proinflammatory cytokines, ameliorate histopathological changes and inhibit mortality in mice rendered septic by LPS administration. Nothing is known about the effect of VIP on production of proinflammatory cytokines in human monocytes infected by virulent Salmonella.
The aim of the current study, therefore, was to investigate the effect of vasoactive intestinal peptide on the production of inflammatory cytokines in human monocytes exposed to S. Typhimurium 4/74. Results: Our finding demonstrates that freshly isolated human monocytes produce proinflammatory cytokines such as TNFα, IL-6, IL-1β and also antiinflammatory cytokines such as IL-4 and IL-10 following bacterial challenge. Conclusion: However, co-culture of infected monocytes with VIP (10-7 M) significantly reduced (P < 0.05) production of TNFα, IL-6, IL-1β but significantly increased (P < 0.05) concomitant production of anti-inflammatory IL-10.

P28
Mortality reduction in patients with severe sepsis and septic shock through a comprehensive sepsis initiative. Results: Prior to implementation of the program (April 2009) through April 2014, ECH achieved a relative reduction of 68% in the mortality rate among those with severe sepsis and septic shock (P = 0.03). This equates to 1,456 lives saved ( Figure 1) [1][2][3]. During the initiative ECH maintained a lower than expected average LOS compared to other hospitals within the Premier database ( Figure 2). Conclusion: The significant and sustained decrease in mortality among those with severe sepsis and septic shock was achieved through the structure and support of multisite collaboration with the INLP and robust internal operations [1,3]. The focus is now on sustainability; including key elements pertaining to accountability, affordability, compassionate care and systematic excellence [4]. Introduction: In 2002 the international Surviving Sepsis Campaign was initiated. Following this, Dutch authorities introduced a nationwide safety campaign, encouraging screening for sepsis and advocating early treatment.
In 2010, the Albert Schweitzer hospital, a large community hospital, introduced a screening program for sepsis in the emergency department (ED) [1]. The goal of this study was to evaluate the bacterial outcome in patients targeted with this campaign. Methods: All patients 18 years and older visiting the ED were screened using the criteria of systemic inflammatory response syndrome (SIRS). Patients with more than two SIRS criteria and a clinical suspicion for infection were eligible for prompt antibiotic administration, after a short assessment by an ED physician. Patient data were collected prospectively, but a retrospective analysis was conducted using a cohort of patients presenting in the ED in the first 6 months of 2011. The definitions for sepsis severity were derived from the guidelines of the Surviving Sepsis Campaign in 2008 and criteria to define bacterial infection were derived from an article by Limper and colleagues [2].
Results: A total of 269 patients were included in the study. Review of infectious outcomes showed no evidence of bacterial disease in 79 (29.4%) patients. Of these patients, 70.9% were in the lowest category of sepsis (SIRS and clinical suspicion of infectious disease). Patients in the lowest category were less likely to suffer from bacterial infection than patients (P = 0.046, see Table 1). In the patients without objective bacterial infection, the median duration of antibiotic treatment was 7 days (IQR 4 to 10). Overall mortality was 7.8 %, which is low compared to current literature regarding (severe) sepsis [3,4], but comparable to literature addressing SIRS and fever in the ED settings [5][6][7][8].
Conclusion: Much effort has been put into promoting early antibiotic treatment for (bacterial) sepsis. However, overtreatment has hardly been addressed and no optimal screening strategy has been identified. Evaluation of our screening protocol using SIRS criteria showed that almost 30% of patients did not suffer from bacterial infection but did receive antibiotic treatment for a median duration of 7 days. Future investigations should address the possible negative effects of overtreatment.   Introduction: Sepsis is the predominant cause of morbidity and mortality in patients with peritonitis [1][2][3][4][5][6]. The Surviving Sepsis Campaign (SSC) is an international effort in reducing mortality based on evidence-based guidelines [7][8][9][10][11][12][13]. This study aims to assess the impact of audit-based feedback in a Plan-Do-Study-Act (PDSA) format on improving implementation of the SSC guidelines in patients with generalised peritonitis at our centre.
Methods: This prospective observational study was conducted in four audit cycles in PDSA format. Multidepartmental inputs were taken to suggest modifications in practice. A questionnaire-based analysis of reasons for noncompliance was done to find out the opinions and reasons for noncompliance. Morbidity, mortality, the ICU and hospital stay among these patients were also analysed.
Results: The baseline compliance with i.v. bolus administration, CVPguided fluids and inotrope supports when indicated were 100%. Over the course of the three audit cycles, statistically significant improvement in compliance was noted for antibiotic administration within 3 hours of presentation (46% to 90%) ( Table 1 Figure 1), obtaining blood cultures before antibiotics (13.8% to 72.5%) ( Table 1 Figure 2) and serum lactate measurement (0% to 78.2%) ( Figure 3). Overall bundle compliance improved from 9.2% to 64.7% (Table 2 Figure 4) by the end of Audit III. The mortality rate decreased from 32.3% to 20% (Table 2 Figure 5).
Conclusion: This study demonstrates that audit-based feedback is a dependable means of improving compliance with SSC guidelines. It brings about improvement by educating users, by modifying their behaviour through feedback and also enhances process improvement by identifying and correcting systemic deficiencies in the organisation. Use of the bundle of care to improve patient outcomes is becoming more widespread; however, their use is more common internationally than in India. Methods: A surveillance study for SSI after routine surgical procedures was conducted from September 2012 until August 2014. A bundle of care consisting of five elements covering the surgical process was introduced in September 2013. The elements of the bundle were perioperative antibiotic prophylaxis, hair removal before surgery, perioperative normothermia, perioperative euglycemia and operating room discipline. Normothermia was defined as a temperature between 36.0 and 38.0°C. Euglycemia was defined as blood glucose <180 mg/dl. Antibiotic prophylaxis was given 15 minutes before the incision. Hair removal whenever needed was done with clippers. Theatre discipline was counted for following points: permanent wearing of scrub suits, surgical cap and mask covering the nose and mouth by all persons in the OR during the surgical procedure. The incidence of SSI was studied as a primary outcome. Morbidity/mortality for next 3 months was studied as the secondary outcome.
Results: Implementing the bundle of care led to a decline in infection rate from 15% before the intervention to 11.4% after the introduction of bundle of care, a fall of 27%, which is significant (P < 0.001). No significant difference in 3-month mortality was found. The compliance to this bundle of care also steadily increased from 10% in September 2013 to 55% in August     Introduction: Fluid bolus therapy (FBT) is a ubiquitous intervention in intensive care. However, the physiological effects in the critically ill are poorly understood. Therefore, we systematically reviewed the contemporary literature to determine the current practice and effect of FBT in the management of severe sepsis and septic shock. Methods: We interrogated the MEDLINE, CENTRAL and EMBASE electronic reference databases using a combination of terms to define a set of records of studies of fluid administration in patients with severe sepsis or septic shock. To achieve contemporary relevance, results were limited to Englishlanguage studies in adults between 2010 and 2013. Results: We identified 22 prospective observational studies, four retrospective observational studies, two quasi-experimental studies, and five randomised controlled trials (RCTs), 41 boluses in total. No RCT compared FBT with alternative interventions. The median fluid bolus was 500 ml (range: 100 to 1,000 ml) administered over 30 minutes (range: 10 to 60 minutes) and 0.9% sodium chloride solution was the most commonly administered. Although 17 studies describe the temporal course of physiological changes after FBT in 31 patient groups, only three studies describe the physiological changes at 60 minutes, and only one study beyond this point (Figure 1). No studies related the physiological changes after FBT with clinically relevant outcomes.
Conclusion: There is a need for obtaining randomised controlled evidence for the physiological effects of FBT in patients with severe sepsis and septic shock beyond the period immediately following its administration.

P35
Forty percent of hospitalizations after severe sepsis are potentially preventable. Introduction: Patients are frequently rehospitalized in the 90 days after severe sepsis. The rate of readmission exceeds patients' baseline rate of hospitalization, and also exceeds the rate after matched nonsepsis hospitalizations [1]. We sought to determine the most common readmission diagnoses after severe sepsis, the extent to which readmissions may be preventable, and whether the pattern of readmission diagnoses differs from that of nonsepsis hospitalizations. Methods: We studied participants in the US Health and Retirement Study with linked Medicare claims (1998 to 2010) [2]. Using validated methods [3,4], we identified severe sepsis and nonsepsis hospitalizations, then measured 90-day readmissions in each cohort. Using Healthcare Cost & Utilization Project's Clinical Classification Software [5], we determined the 10 most common readmission diagnoses after severe sepsis. We measured rates of 'potentially preventable' readmissions using published definitions [6]. We compared rates of all-cause, potentially preventable, and causespecific hospitalizations between survivors of severe sepsis and nonsepsis hospitalizations using chi-squared tests. Results: We identified 3,703 severe sepsis and 44,840 nonsepsis hospitalizations, of which 3,036 (82.0%) and 43,539 (93.1%) survived to discharge, respectively. In the next 90 days, 43.6% of severe sepsis survivors were rehospitalized, compared to 34.8% of nonsepsis survivors, P < 0.001. The top readmission diagnoses following severe sepsis (Table 1) included several recognized potentially preventable diagnoses: heart failure, pneumonia, exacerbation of chronic obstructive pulmonary disease (COPD), and urinary infection. Also common were readmissions for sepsis, acute renal failure, and aspiration pneumonitis, diagnoses that could plausibly be prevented or treated early to prevent hospitalization. Patterns of readmission differed in severe sepsis survivors; rates of readmission for sepsis, renal failure, respiratory failure, device complication, and aspiration pneumonitis were higher and accounted for a greater proportion of the total readmissions. Potentially preventable hospitalizations -infection (sepsis, pneumonia, urinary tract, and skin or soft tissue), heart failure, COPD exacerbation, acute renal failure, and aspiration pneumonitisaccounted for 40.5% of all readmissions after severe sepsis (compared to 25.8% following nonsepsis admission, P < 0.001), and 19.6% of severe sepsis survivors experienced a readmission for one of these diagnoses (compared to 9.5% following a nonsepsis admission, P < 0.001).
Conclusion: Forty percent of hospitalizations after severe sepsis occur for diagnoses that may be preventable. A few disease categories account for a relatively large proportion of the hospitalizations after severe sepsis, suggesting the feasibility of tailoring postdischarge interventions to patient's personalized risk for these common postsepsis diagnoses.   Introduction: Enterococcal bacteraemias are associated with high mortality rates, ranging from 23 to 48% in critically ill patients. However, it remains uncertain whether these events are causes or merely markers of disease severity and mortality. Our aim was to describe the epidemiology, management and clinical outcomes of ICU-acquired enterococcal bacteraemia in comparison to other frequently isolated pathogens. Furthermore, we aimed to estimate the population attributable fraction of ICU mortality caused by enterococcal bacteraemia.
Methods: Between January 2011 and March 2013 we included consecutive patients with an ICU length of stay of at least 3 days in two tertiary care centres in the Netherlands. ICU-acquired bacteraemia was defined as a first positive blood culture occurring at least 3 days after ICU admission. Enterococcal bacteraemias were compared to other frequently isolated pathogens with respect to patient characteristics, their management and outcomes. We used competing risk survival regression, a multistate model and cumulative incidence functions to estimate the population attributable fraction of ICU mortality due to enterococcal bacteraemia.
Results: Out of a total of 3,108 patients, 222 (7.1%) patients were responsible for 272 events of ICU-acquired bacteraemia, of which 76 were due to enterococci, 124 to coagulase-negative staphylococci and 40 to Gramnegative bacteria. Patients with enterococcal bacteraemia were more severely ill compared to patients with bacteraemia caused by other pathogens. In comparison to patients with coagulase-negative staphylococci, those with enterococci were more frequently managed with renal replacement therapy and also had more intravascular or orthopaedic hardware. Although crude ICU mortality was higher in patients with bacteraemias due to enterococci compared to coagulase-negative staphylococci, this association disappeared after adjustment for confounders (subdistribution hazard ratio 1.04, 95% confidence interval 0.65 to 1.68). The population attributable fraction of ICU mortality due to enterococcal bacteraemia was 4.7%. Conclusion: Bacteraemias with enterococci occur in more severely ill patients, but their virulence seems comparable to that of coagulasenegative staphylococci. Furthermore, the population attributable fraction of ICU mortality due to enterococcal bacteraemia is low. Therefore, enterococcal bacteraemias are more likely to be markers than causes of increased disease severity and mortality. Introduction: Pneumonia is the major cause of sepsis, responsible for almost one-half of all sources of infection [1]. Sepsis and septic shock lead to organ failure and death. Spreading of microorganisms and their toxins through the blood could contribute to the organ dysfunction. The heart, liver and kidney are examples of organs damaged during the systemic infection and organ failure predicts poor prognosis in patients with sepsis [2]. However the correlation, if any, between bacterial spreading and organ injury is unclear. Thus, the aim of this study was to study the bacterial systemic spreading from a localized infection along with time and the target organ inflammatory profile using proinflammatory cytokines as surrogate markers.
Methods: Pneumosepsis was induced by Klebsiella pneumoniae as described in [3]. The number of viable bacteria inoculated was 10 9 colonyforming units to achieve a mortality rate of~50% by 48 hours. Blood, lung, heart, liver, spleen, kidney and brain were aseptically harvested, homogenized and plated on Müeller-Hinton agar dishes, for 18 hours at 37°C. Plasma and the tissues homogenates were assayed for interleukin-6 (IL-6) and interleukin-1β (IL-1β) using commercially available enzymelinked immunosorbent assay kits according to the manufacturer's recommendations (PeproTech Inc., Rocky Hills, NJ, USA). Results: Lung inoculation with K. pneumoniae evolved to systemic spreading of bacteria to all organs, mostly the liver and kidney. Surprisingly, bacteria were found as early as 30 minutes in vital organs such as the brain and heart. The infection in the organs rose steadily up to 24 to 48 hours. Significant increases in IL-6 and IL-1β were found in the plasma, 24 hours after infection. However, cytokine levels in the organs were as high as fivefold the plasma levels.
Conclusion: Our data show that the early bacterial dissemination may be important for the onset of organ inflammation. Assuming that the higher organ cytokine level is a marker of an ongoing inflammation, this may explain organ dysfunction during sepsis. Thus, our study suggests that systemic (meaning blood) parameters may not reflect the severity of inflammation/dysfunction in target organs, and that might be the determinant to sepsis outcome.

P40
Effects of splenectomy and GTS-21, a selective a7 nicotinic acetylcholine receptor agonist, on the development of septic ileus in mice. Introduction: Sepsis remains a leading cause of mortality in our ICUs. Ileus, defined as the inhibition of the propulsive motility in the gastrointestinal (GI) tract, together with mucosal barrier dysfunction will maintain sepsis by the translocation of intestinal bacteria. Preliminary data in our group showed that the administration of GTS-21, a selective alpha7 nicotinic acetylcholine receptor (α7nAChR) agonist, ameliorates inflammation and GI motility disturbances during sepsis [1]. Activation of the α7nAChR is the final step in the vagal anti-inflammatory pathway, a spleen-dependent and macrophagedependent pathway that dampens inflammation. We aimed to study the effects of splenectomy (SPLX) combined with GTS-21 on GI motility and on local colonic and systemic inflammation, as the role of the spleen in the vagal anti-inflammatory pathway is currently under debate. Methods: Septic ileus was induced in Swiss-OF 1 mice by cecal ligation and puncture (CLP). One week prior to CLP, mice underwent SPLX or sham surgical incision (sham). GTS-21 (8 mg/kg) or vehicle was administered intraperitoneally 1 hour before CLP, and once every 24 hours after the procedure. Mice were sacrificed 48 hours following CLP. Motility was assessed by measurement of GI transit of beads, with calculation of percentage of gastric emptying (%GE) and geometric center (GC). Serum samples were analyzed with cytometric bead array (CBA) for the proinflammatory cytokines IL-6 and TNFα. Supernatants of centrifuged homogenized colonic samples were assessed with CBA for IL-6 and TNFα Results: Administration of GTS-21 resulted in a significant improvement in GI motility in septic mice, as was demonstrated by an increase in the %GE and GC (Table 1). GTS-21 significantly decreased the serum concentration of IL-6, but not TNFα, in septic mice treated with GTS-21 compared to vehicle-treated mice. Prior SPLX protected mice from developing ileus following CLP. SPLX + vehicle mice demonstrated a significantly lower serum concentration of IL-6 and TNFα compared to sham + vehicle, and colonic TNFα levels declined significantly following SPLX. Administering GTS-21 in SPLX mice did not result in an additional benefit on GI motility, nor on systemic or local colonic inflammation (Table 1). Conclusion: Splenectomy protects mice from developing ileus following CLP, an animal model of polymicrobial sepsis, as does the preventive administration of the selective α7nAChR agonist GTS-21. Splenectomy resulted in a reduction of systemic TNFα and IL-6 levels, indicating that the spleen is a major source of proinflammatory cytokines. GTS-21 ameliorates CLP-induced septic ileus, but did not lead to a complimentary benefit in addition to splenectomy.
Introduction: Biomarker discovery research has not focused on the emergency department (ED) due to perceived lack of access to ED patients for study, difficult patient identification, and preemption by time-critical clinical needs. The aim of this study was to use an automated process to accrue, within the ED, a biobank of patients presenting at risk for severe sepsis.
Methods: Our group has previously derived an algorithm for identification of at-risk patients of which about 45% will be severely septic [1]. Here, we utilize that algorithm to identify a prospective cohort to be included in our ED sepsis biobank. Patients are excluded if they are pregnant, <18 years old, a vulnerable adult, or imprisoned. An electronic notification system identifies patients and automatically pages phlebotomy and laboratory processing personnel directly. Blood draws occur immediately after patient identification and at 6 and 24 hours into hospitalization. We use an IRBapproved delayed consent model that minimizes the time between identification and blood draw. Plasma extracts and cellular isolates are processed and stored immediately. Clinical data are extracted for relevant patient outcomes, severe sepsis, and indicators of critical illness. Results: In 1 year of accrual, the alert algorithm identified 1,773 eligible patients for screening. Blood was drawn from 873 patients based on phlebotomy availability and laboratory processing staffing. A total of 642 patients completed blood draws and consented for inclusion, with the balance of patients having been discharged, transferred, or deceased with no identifiable legally authorized representative for consent. The median time from alert to first blood draw was 25 minutes (range 5 to 60 minutes).
In an observation period, during which 227 consecutive patients had blood drawn, only 12 patients or families (5.2%) declined consent after the initial blood draw. Conclusion: We have successfully developed and implemented a biobank in the ED that functions within practice limitations and has high patient accrual. Automation is key for efficient implementation in the ED. Using automated design improvements, we obtain blood samples from sepsis patients presenting to the ED faster than previous studies. Prompt biobank collection within the ED allows research into mechanisms of sepsis earlier in a patient's hospital course than has heretofore been possible and enables biomarker discovery for testing development directly relevant to emergency providers. Introduction: Fever predicts an infection cause of SIRS/sepsis more specifically than other commonly used hemodynamic criteria. Traditionally, a body temperature of 38.0°C has been used to define fever, but this cutoff is based on limited research. Furthermore, it has been proposed that fever responses are blunted in older adults, limiting the utility of fever in infection. This study determines the likelihood that a body temperature will predict diagnosis of infection in the emergency department (ED).
Methods: This was a retrospective cohort analysis of adult patients (>18 years old) presenting to a large academic emergency department from September 2010 to December 2012. Patient age, emergency physician diagnosis, final disposition, and initial body temperature were examined for each patient. Likelihood of a diagnosis of infection was calculated for all temperature ranges. Sensitivity and specificity of fever thresholds for a  Figure 1). In adults of all ages, temperature >38.0°C had a specificity >99% for a diagnosis of infection in the ED although sensitivity was relatively poor. Specificity decreases only slightly with lower fever cutoffs (37.2 to 37.9°C) while sensitivity is increased to a greater degree (see Table 1). Conclusion: The likelihood of a diagnosis of infection increases at the extremes of body temperature and is higher overall among older adults (≥65 years old). In our population, elevated temperature was at least as sensitive for a diagnosis of infection in older people as in younger adults. The use of body temperature for predicting diagnosis of infection has an overall low sensitivity but high specificity. Changing the current practical definition of fever from ≥38.0°C to ≥37.5°C significantly increases sensitivity of predicting infection without greatly impacting the specificity. Only 2.3% of all patients had a temperature between 37.5 and 37.9°C, and these patients were four times more likely to have a diagnosis of infection than those with temperature of 36.0 to 36.9°C.

P43
Ghrelin: an anti-inflammatory theurapeutic agent in septic rats. Introduction: Sepsis is a life-threatening systemic inflammatory syndrome (SIRS), which affects many organ systems, that leads to hemodynamic changes in the presence of suspected or proven infection, advancing to organ dysfunction and failure [1][2][3]. In recent studies, 377 out of 100,000 cases of sepsis were observed while $14,600,000,000 has been determined as the average annual hospital costs. Although there are many studies, the molecular mechanism is not yet clearly elucidated [2,3]. Lipopolysaccharide (LPS) is the lipid molecule which the outer membrane of Gram-negative bacteria used in designing the experimental sepsis model [4]. Ghrelin was discovered in 1999 as a specific ligand for growth hormone secretagogue receptor and then pleiotropic effects such as anti-inflammatory, antioxidant, and so forth were found. Ghrelin was released in many tissues and organs in which there also was its receptor. The liver is an organ which has ghrelin receptors and was affected by sepsis primary [5,6].  (5 mg/kg, ghrelin 10 nmol/kg i.v., n = 10). Rats were decapitated 24 hours after first injection. We aimed in this study to investigate effects of ghrelin in sepsis which is created by LPS with sepsis descriptive parameters such as body temperature and leukocyte count with proinflammatory cytokine TNFα and anti-inflammatory cytokines IL-10 by ELISA, and hematoxylin and eosin stain for observed morphological changes. Results: We detected the increase of leukocyte number, hypothermia, proinflammatory and anti-inflammatory cytokines such as TNFα and IL-10 as developing inflammatory reactions, resulting in hemodynamic and metabolic changes in rats treated with LPS. Also in groups with ghrelin treatment, ghrelin affects leukocyte numbers, body temperature, proinflammatory and anti-inflammatory cytokine levels and histological changes in controls and the LPS group (Figures 1 to 4). As a result of histologic examination, the curative effect of ghrelin partially on liver tissue damage is observed ( Figure 5).

Conclusion:
We think the results of ghrelin may be affected depending on the dose and duration. Also partial healing effects of ghrelin in our results on this topic at the molecular level will contribute to other studies.   Introduction: Blood sugar control for patients with sepsis remains controversial. We aimed to test the hypothesis that the variation of blood sugar level is associated with patient outcome in this study.
Methods: A retrospective cohort study on nontraumatic adult patients who visited the ED of a tertiary hospital in 2010 and had a clinical diagnosis of severe sepsis was conducted. Patients with two sets of blood culture ordered by emergency physicians and at least two blood sugar tests results available during the first 48 hours of hospitalization were included. The coefficients of variation (CoV, the ratio of the standard deviation to the mean) of the blood sugar level were analyzed with multivariate logistic regression models to test the association between inhospital mortality.
Results: Of the 1,537 patients included, most were older than 70 years of age (median; 71, IQR: 59 to 80), male (54%), without a diagnosis of severe sepsis (63%) and had a previous diagnosis of diabetes (84%). The initial blood sugar levels of patients with and without previously diagnosed diabetes were 259 ± 9.9 and 154 ± 5.7, respectively (mean ± SEM). The CoV of the consecutively monitored blood sugar level during the first 48 hours of admission for patients with and without previously diagnosed diabetes were 29.4 ± 0.5% and 21.0 ± 0.5%, respectively. Patients with CoV lower than 10% and higher than 30% tended to have higher mortality rate, compared to patients with 10 to 30% CoV level (11% vs. 12% and 7%, respectively, Figure 1). In the multivariate logistic regression model adjusting for age, initial blood sugar level, severity of sepsis, previous diagnosis of diabetes and diagnosis of severe sepsis, the higher CoV level (>30%) was found to be associated with 60% increased odds of in-hospital mortality (aOR: 1.61 ± 0.34); while the previous diagnosis of diabetes was found to be associated with 45% lower odds of in-hospital mortality (aOR: 0.55 ± 0.13, Table 1). Conclusion: In this retrospectively cohort study, we found that increased blood sugar variation was associated with worse patient outcome. However, further study is merited to test the possible causal relationship between variation of blood sugar level and patient outcome.

P46
Association of initial intracellular signalling pathway and cytokine level with early mortality in severe sepsis patients. Excessive inflammatory process with maladaptive host's immune response leads to organ dysfunctions and death [1,2]. This study was designed to investigate association of the initial level of intracellular signalling pathway and cytokine involved in sepsis pathophysiology (that is, IKK-β, NF-B, tumour necrosis factor (TNF)α) and 72-hour (early) mortality in severe sepsis patients.
Methods: A prospective cohort study was conducted in severe sepsis patients (aged 18 years and older) admitted to the Emergency Unit of Cipto Mangunkusumo Hospital, Persahabatan Hospital, and Gatot Subroto Indonesia Central Army Hospital, Jakarta, Indonesia. All blood samples for intracellular signalling pathway (that is, IKK-β, total NF-B, phospho NF-B) and cytokine (that is, TNFα) were collected and measured using the ELISA method during first 24 hours of inclusion. Patients' outcome was observed during first 72 hours after inclusion. Mann-Whitney test was used to analyse the median difference of IKK-β, total NF-B, phospho NF-B level in two groups based on 72-hour survival. The t test was used to analyse the mean difference of TNFα levels in both groups.
Results: Subjects consisted of 90 patients. Early mortality developed in 27 subjects. Baseline characteristics of survival and death subjects are shown in Table 1. The initial intracellular signalling pathway and cytokine parameters level are shown in Table 2. There was a significant higher median first 24 hours IKK-β and total NF-B level in survival subjects compared with death subjects (P = 0.025 for median IKK-β and P = 0.036 for median total NF-B). There was no significant difference of initial phospho NF-B and TNFα level in survival and death subjects.
Conclusion: There is a significant lower initial IKK-β and total NF-B level in severe sepsis patients surviving on 72-hour observation. There is a tendency of lower initial phospho NF-B and TNFα level in severe sepsis patients surviving on 72-hour observation. Introduction: The pre-existing inflammatory state influences the behavior of infection, a first peak of inflammation may be the risk factor during infection and predispose to a severe form of the disease [1,2]. This hypothesis is frequently accepted but has never been tested. Patients with hematological malignancies treated by chemotherapy in case of autologous stem cell transplantation (ASCT) are exposed to a neutropenic state: an abnormal low number of neutrophils. In the neutropenic patient, sepsis is more frequent than in the non-neutropenic patient. The severity of infection depends on many factors, particularly the pre-existing inflammatory state, the genetic factors, and the duration and the depth of neutropenia. In the literature, the factors involved in inflammation and infection are often confused. On the other hand, no predictive transcriptomic signature of sepsis was found. Our objective is to identify predictive genes that influence the outcomes of the infection in neutropenic patients. Methods: To test our hypothesis, high-throughput transcriptomic analysis and bioinformatics will be combined, laying on the modulation of gene expression of the peripheral blood mononuclear cells of patients with hematological malignancies treated by chemotherapy in case of ASCT.
Results: A preliminary study on a reduced number of patients has allowed us to assess the feasibility of the project. After a statistical analysis, 615 genes are differently expressed between patients who developed sepsis and those who have not developed with a false discovery rate of 5%. These 615 genes are potentially predictive of sepsis 2 days before the development of the infection. Furthermore, 28 genes are differentially expressed between patients who developed sepsis and those who have  not developed with a false discovery rate of 5%, and these 28 genes are potentially predictive of sepsis 7 days before the development of the infection.
Conclusion: The ongoing transcriptomic study with a larger effect will allow us to test our hypothesis and confirm the results of our preliminary study. The microarray results should be strengthened with quantitative PCR assays, than validated with protein assays. Finally, we will transfer the experimental data to clinical practices.

P48
Histopathological changes in septic acute kidney injury in critically ill children: an observational analytical study of postmortem renal biopsies. R Rameshkumar 1* , S Mahadevan 1 , RN Ganesh 2 , P Narayanan 1 , V Bhat 1 Introduction: Critically ill children often manifest acute kidney injury (AKI) as part of their disease process with incidence up to 82% and it is independently associated with poor outcome [1]. Sepsis is the most common contributing factor and little is known about the pathogenesis of septic AKI. There are no consistent histopathological changes in human or experimental septic AKI [2]. Hence, understanding of the structural changes associated with its occurrence is therefore important in the management of AKI in critically ill children. Methods: Children aged less than 12 years who died with septic AKI were screened for percutaneous USG-guided kidney biopsy after obtaining written informed consent from July 2012 to June 2014. Three cores of kidney tissue were taken for histopathological evaluation by light, immunofluorescence and electron microscopic examination. Sepsis and AKI was defined using international pediatric sepsis consensus conference and pRIFLE criteria respectively. Events related to death, laboratory parameters and microbiological details 24 hours preceding death were recorded.
Results: A total of 40 kidney biopsies were done during the study period. Median (IQR) age was 12 (2 to 36) months and PRISM-III was 14 (12 to 18). The most common change was normal histology in 37.5% (n = 15) followed by tubular change alone in 22% (n = 10), glomerular change alone in 5% (n = 2) and blood vessel change in 2.5% (n = 1) of specimens. Twelve specimens were showing a combination of changes (tubular + glomerular + interstitium = 5; tubular + glomerular + blood vessels = 3; tubular + glomerular = 2; tubular + interstitium = 1; tubular + glomerular + interstitium + blood vessels = 1). All tubular changes were consistent with acute tubular necrosis (ATN) and changes involved from 5 to 15% of cortex. Thrombotic microangiopathy was diagnosed in 12.5% (n = 5) of specimens. There were no significant difference between ATN versus non-ATN groups with respect to baseline characteristics and events related to mortality. Conclusion: The most common change in septic AKI in critically ill children is normal histology followed by ATN. This is consistent with published literature [2]. Introduction: Healthcare-associated infections (HAI) are a significant problem in the pediatric intensive care unit (PICU). Apart from contributing to mortality, they also increase the coast of PICU care [1]. Surveillance and prevention of HAI by multifaceted quality improvement intervention among critically ill children remain part of the standard of care [2]. Methods: The study was conducted in the 19-bed PICU of a tertiary care referral academic institute. Data regarding ventilator-associated pneumonia (VAP) and central line associated bloodstream infection (CLABSI) using the CDC definition were collected prospectively from July 2013 to June 2014. Multifaceted quality improvement intervention consisting of infection control nurse and physician and hand hygiene education module and wearing a gown and mask during the care of critically ill children was introduced from January 2014. Incidence of VAP and CLABSI was compared before (July to December 2013) and after (January to June 2014) introduction of the intervention. Results: Before the intervention period, the incidence of VAP was 28.5 per 1,000 ventilation-days and CLABSI was 13.7 per 1,000 catheter-days. After the intervention period the incidence of VAP was 13.3 per 1,000 ventilation-days and CLABSI was 8.3 per 1,000 catheter-days. The proportion of patients ventilated for more than 48 hours who had VAP was significantly less after intervention as compared to before the intervention period (14.2%, n = 25/176 vs. 25.2%, n = 29/155; P = 0.012, odds ratio (OR), 95% CI 0.49, 0.28 to 0.86). Both groups were similar with respect to age, sex ratio, severity (PRISM-III), device utilization rate and grade of infection. No significant difference occurred in overall PICU mortality before and after intervention (28.2% vs. 28.9%).
Conclusion: Multifaceted quality improvement intervention results in significant reduction of the healthcare-associated infection rate although it was higher than reported from developed countries [1][2][3].
Blood cultures were also evaluated and processed by a BacT/Alert 3D system (bioMérieux, Italy); CRP was measured by immunonephelometry (Siemens Healthcare Diagnostic, Italy) and PCT was assayed by an enzyme-linked fluorescent assay (VIDAS BRAHMS PCT, bioMérieux, Italy); presepsin levels were measured by rapid automated PATHFAST immunoanalyzer (kindly provided by GEPA SRL, Italy), based on chemiluminescent enzyme immunoassay. A statistical analysis was carried out by Mann-Whitney test.
Results: Presepsin and PCT levels were significantly higher in culturepositive subjects versus negative controls; such difference was found even at the admission time. The presepsin values in worsening/dead patients exhibited a significantly higher level at admission time. On the contrary, in the same group of patients, PCT exhibited a decrease of its level. In poor prognosis patients CRP showed a quite irregular kinetic, although in such a group the admission value was higher than the same marker in live subjects.
Conclusion: In this preliminary study, presepsin and PCT levels exhibited substantial higher values in culture-positive patients. The kinetic curves of presepsin, obtained from both survival and worsening/dead subjects, revealed the optimal performance of this biomarker, particularly in severely ill patients, as also shown in other studies. During sepsis, increase of presepsin levels may be a more reliable marker, indicating an unfavorable outcome [3,4]. Furthermore, high presepsin levels could alert clinicians not to suspend antibiotic treatments even after clinical symptoms have improved and PCT levels have returned to normal. Introduction: T-cell depletion is a marker of the hypo-inflammatory phase of sepsis [1,2]. Systemic T-cell loss could be caused i.a. by thymus involution during sepsis, thus contributing to immune paralysis [3]. Usually, thymic T-cell egress is mediated by a sphingosine-1-phosphate (S1P) gradient. That is, T cells leave the thymus towards an increased S1P-level in the periphery (blood/lymph) [4]. A disruption of this S1P gradient blocks T-cell emigration. S1P is produced by sphingosine kinases 1 and 2 (SPHK1 and SPHK2), which are ubiquitously expressed [5]. Apoptotic cells are known to release S1P [6]. We suggest that apoptotic events in the thymus during sepsis increase S1P levels, thereby disrupting the gradient and preventing egress of T cells from the thymus. Methods: We used a murine polymicrobial sepsis model to analyze thymus involution. Sepsis is induced in wildtype mice, SPHK1 -/-, or SPHK2 -/mice by cecal ligation and puncture (CLP). Thymus involution was determined by analyzing the T-cell amount of single-positive mature (CD4 + CD8vs. CD4 -CD8 + ), double-positive late immature (CD4 + /CD8 + ), and double-negative early immature cells by FACS analysis. T cells from SPHK1 -/or SPHK2 -/mice, which produce less S1P, and wildtype mice treated with a SPHK1 inhibitor were used to characterize whether T cells in these mice have a higher rate of emigration compared to control mice. Thymic and serum S1P levels were quantified by LC-MS/MS. Apoptosis was determined by annexin V FACS staining. SPHK mRNA levels were determined by qPCR.
Results: The thymus of septic mice showed more CD3-positive cells but a decreased number of CD4/CD8 double-positive T cells, pointing to a thymic retention of single-positive mature T cells. In line with our assumption, CLP causes apoptosis of thymocytes and increases SPHK1 mRNA expression.
Concomitantly S1P levels are increased in the thymus and consequently decreased in serum following CLP. The knockout of the S1P producing sphingosine-kinases SPHK1 or SPHK2 and the pharmacological inhibition of SPHK1 restores T-cell egress as indicated by an increase of double-positive immature T cells compared to wildtype mice. Conclusion: Our data suggest that inhibition of SPHK1-mediated S1P generation during sepsis restores thymic T-cell egress, which might improve septic outcome. Therefore, understanding mechanisms of thymus involution during sepsis may help to reconstitute thymic function, finally improving immune reactions in the hypoinflammatory phase of sepsis.

P52
Activation of the peroxisome proliferator activated receptor g counteracts sepsis-induced T-cell cytotoxicity towards alloantigenic target cells. Introduction: Sepsis originates from an uncontrolled inflammatory response. Despite intensive research, sepsis remains a major cause of death in ICUs. Therefore, new therapeutic approaches are mandatory. Taking into account that during sepsis progression cytotoxic T cells (CTL) are activated in an autoimmune fashion contributing to multiorgan damage, it remains unclear whether CTL are activated towards alloantigenic cells as well. This is especially important for patients receiving an immune suppressive therapy to permit organ transplantation and thus known to be at high risk for developing sepsis. Therefore, we analyzed whether sepsis activates CTL towards alloantigenic target cells and whether this can be inhibited by PPARγ activation, known to block T-helper cell responses. Methods: To characterize whether sepsis activates CTL and whether this can be inhibited by PPARγ activation, we used an ex vivo cytotoxicity assay to analyze CD8 + T-cell-dependent cytotoxicity. Responder CD8 + T cells were isolated from C57Bl/6N PPARγ wildtype (PPARγfl/fl) and T-cell specific knockout (Tc-PPARγ -/-) mice (haplotype H2Kb) following cecal ligation and puncture (CLP) versus sham treatment. P815 mastocytoma cells, a cell line originally derived from DBA/2 mice (haplotype H2Kd), were used as alloantigenic target cells. Pharmacological inhibition and/or activation of PPARγ in vivo and ex vivo was performed to clarify the impact of PPARγ in blocking CTL-dependent cytotoxicity. In vivo, PPARγ activity in wildtype mice was pharmacologically inhibited by the irreversible antagonist GW9662 or induced by the thiazolidinedione rosiglitazone. Systemic application of both compounds was performed intraperitoneally. A classic splenocyte-driven stimulation protocol to activate CTL was carried out as control.
Results: CTL isolated from septic mice showed enhanced cytotoxicity towards alloantigenic P815 target cells. Enhanced cytotoxicity was effectively reduced by both PPARγ activation in vivo and ex vivo. In line, in CTL isolated from T-cell-specific PPARγ knockout (Tc-PPARγ -/-) mice PPARγ activation was ineffective, strengthening a PPARγ-dependent mechanism. At the molecular level in vivo and ex vivo activation of PPARγ reduced Fas and granzyme B expression in activated CTL, which might explain reduced cytotoxicity. Conclusion: Our study therefore suggests PPARγ activation in vivo to attenuate CTL-dependent alloantigenic cytotoxicity to possibly inhibit acute organ rejection.

P53
Using process mapping to identify barriers to effective management of sepsis in a cancer hospital: lessons for successful implementation of a whole hospital pathway. Introduction: Infection and sepsis are common problems in cancer management affecting up to 45% of patients. However, international guidelines focus on the management of neutropenic fever, and fail to address the recognition and resuscitation of patients who meet sepsis criteria. Peter MacCallum Hospital is a 100 inpatient-bed tertiary cancer hospital with hematology, medical oncology, cancer surgery and radiation oncology, as well as a medical and chemotherapy day unit, apheresis, and large ambulatory service but no emergency department. Records showed that up to 25% of all in-hospital Medical Emergency Team (MET) calls were attributable to sepsis with in-hospital mortality rates of up to 25%. We aimed to identify barriers to effective management of inpatient sepsis at Peter MacCallum Cancer Centre and to implement a hospital-wide sepsis pathway.
Methods: A sepsis working party was formed with the antimicrobial stewardship team, clinicians, and senior ambulatory and inpatient nurses. Each member undertook direct observation and focus group interviews in an allocated clinical area. Three key areas were examined: issues relating the identification of sepsis, issues relating to clinical review of the patient, and issues relating to timely administration of first dose of antibiotic. Inpatient and outpatient issues were graphically represented in a process map (Figure 1). Results: Process mapping revealed significant gaps in knowledge in medical and nursing staff and structural barriers to rapid resuscitation of patients. There were significant knowledge gaps in the awareness of sepsis diagnostic criteria, the role of lactate, effective fluid resuscitation, and the need for early clinical review and referral to the ICU. Examples of structural barriers to effective resuscitation included lack of availability of nurse cannulators, availability of antibiotics, rapid intravenous fluid infusers and sufficient after-hours medical and ICU liaison support. Knowledge and structural barriers were systematically addressed during the implementation of the clinical sepsis pathway. The sepsis pathway was designed as a medical record form to be used across all clinical areas, and supports nurse initiation. Following pathway implementation in March 2013 there have been substantial improvements in the number of patients who have had a lactate taken, received appropriate fluid resuscitation and time to first dose of antibiotics. Administrative data shown in Figure 2 demonstrate increased ascertainment of cases, and a fall in all-cause sepsis mortality after the pathway was commenced.
Conclusion: Identifying knowledge gaps and structural barriers using process mapping led to the successful design and implementation of a sepsis program. The figures show that despite increased rates of coded sepsis cases, mortality and ICU admission rates have not increased. Mortality Introduction: Previous anti-inflammatory strategies against sepsis, a leading cause of death in hospitals, had limited efficacy in clinical trials, in part because they targeted single cytokines and the experimental models failed to mimic clinical settings [1][2][3]. Neuronal networks represent physiological mechanisms, selected by evolution to control inflammation, that can be exploited for the treatment of inflammatory and infectious disorders [3]. Methods: Animal procedures were approved by the Institutional Animal Care & Use Committee of the New Jersey Medical School of Rutgers University. All animal experiments were performed in 6-week-old to 8-weekold (~25 ± 5 g) male mice without any exclusion criteria. Experimental sepsis: endotoxemia and CLP were performed as we previously described. LPS was dissolved in sterile pyrogen-free PBS and sonicated for 30 minutes immediately before use. Mice received a LD 50 dose of LPS (6 mg/kg body weight i.p.). LPS was added to the whole blood to a final concentration of 250 ng/ml for the in vitro procedures. Selective neurectomies and electrical stimulations: all selective neurectomies and electrical stimulations were performed in mice anesthetized with ketamine and xylazine. The electrical stimulation in electroacupuncture and direct nerve stimulation (sciatic and vagus nerves) was performed with a continuous-mode stimulation for 15 minutes with a electrical potential difference of 4 V, an electric current of 40 mA, a pulse width of 50 μs and a frequency of 10 Hz using an electrostimulator.
Results: Here, we report that sciatic nerve activation with electroacupuncture controls systemic inflammation and rescues mice from polymicrobial peritonitis. Electroacupuncture at the sciatic nerve controls systemic inflammation by inducing vagal activation of aromatic L-amino acid decarboxylase, leading to the production of dopamine in the adrenal medulla. Experimental models with adrenolectomized mice mimic clinical adrenal insufficiency [4], increase the susceptibility to sepsis and prevent the anti-inflammatory effects of electroacupuncture. Dopamine inhibits cytokine production via dopamine type 1 (D1) receptors. D1 receptor agonists suppress systemic inflammation and rescue mice with adrenal insufficiency from polymicrobial peritonitis. Our results suggest a new anti-inflammatory mechanism mediated by the sciatic and vagus nerves that modulates the production of catecholamines in the adrenal glands. Conclusion: From a pharmacological perspective, the effects of selective dopamine agonists mimic the anti-inflammatory effects of electroacupuncture and can provide therapeutic advantages to control inflammation in infectious and inflammatory disorders. Preliminary results in human clinical trials indicate that electroacupuncture attenuates the postsurgical inflammatory response decreasing the serum levels of inflammatory cytokines.

P55
Immunomodulation and infection: identification of small molecule TLR3 blockers to combat deleterious inflammation in pneumonia. Introduction: Pneumonia is a common cause of death worldwide and the leading cause of sepsis and shock in the ICU. Controlling excessive immune stimulation with deleterious consequences for the host organs is a major strategy in severe infections. In addition to viral dsRNA, the immune Toll-like receptor 3 (TLR3) senses RNA released from injured tissues and necrotic cells promoting excessive inflammatory cytokine release in pulmonary epithelial, endothelial and immune cells. Thus, as a potent regulator of the lung immune response, TLR3 represents a specific target to control damaging inflammation in severe lung infections in combination with antimicrobials.
Here, we aimed to characterize potent small molecules displaying in vitro and in vivo TLR3 blocker activities and develop several preclinical models to investigate the role of TLR3 in pneumonia.

Results:
We identified compounds with anti-TLR3 activity (IC 50 = 50 nM) through random screening and structure-activity relationship analysis by testing small molecule libraries on recombinant hTLR3-HEK293 cells. Selected compounds interacted with mouse and human TLR3, were devoid of effect on TNFα-induced activation at 1 µM on HEK293 cells and exhibited good early-ADME properties. Hit molecules also counteracted PolyAU (PAU)/Poly(I:C) (PIC)-induced cytokine release (IL6, IL8, IP-10) on human bronchial epithelial cells (BEAS-2B) expressing native TLR3 receptors. Interestingly, sustained stimulation of BEAS-2B by PAU/PIC (0.1 to 10 µM) to mimic viral activation induced overexpression of TLR3 mRNA, which was dose-dependently inhibited by these compounds, as evidenced by RT-qPCR. In vivo, we confirmed the anti-TLR3 activity (3 to 30 mg/kg i.p.) against PAU/PIC (100 µg/mouse i.v.) but not flagellininduced plasma cytokine release in CD1 mice. To study these TLR3 inhibitors under pathological situations, we developed models of Streptoccocus pneumoniae (ATCC6303) or Pseudomonas aeruginosa (PAO-1)-induced pneumonia post influenza A virus (IAV PR/8/34 H1N1) infection. Mice previously challenged with intranasal IAV showed enhanced susceptibility to lung bacterial infections with a dramatic mortality rate, whereas separately these pathogens are not lethal. In addition, IAV/S. pneumoniae co-infected mice showed increased lung and blood bacterial loads with severe inflammation signs evidenced by elevated systemic IL-6 and KC levels and important lung tissue damages. Currently, TLR3 blockers are ongoing evaluation in these models. Conclusion: Immunomodulation and personalized treatments are becoming relevant approaches to manage severe infections. Here, we discovered the first small molecule probes targeting mouse and human TLR3. These compounds efficiently control TLR3-triggered proinflammatory response in vitro and in vivo and shape TLR3 overexpression in human lung epithelial cells. They represent valuable pharmacological tools to study the contribution of TLR3 in pneumonia and to propose new adjuvant immunotherapies.
Introduction: Acute kidney injury (AKI) is a frequent and serious complication of sepsis. Moreover, there is strong evidence that AKI in patients with severe sepsis is associated with a higher mortality rate. The     Figure 3). In the LPS + Simvastatin group, we found decreased activity of complex I compared with those of the LPS and Simvastatin groups (P = 0.05, P = 0.07, respectively). As a result of the light microscopic examination with H&E stained sections, we observed tubule lumens widened and partially damaged in the epithelium. In the Simvastatin group was seen partially widened tubular structure and even in some areas tubular structure was found the same as control sections. Moreover, the damage in tubular width and proximal epithelium was observed to continue ( Figure 4). Introduction: Muscular loss is a characteristic phenomenon induced by a massive inflammation as occurring in sepsis. Indeed, the early phase of sepsis is responsible for the expression of proinflammatory cytokines, namely TNFα and IL-6, known to be effectors of cachexia. Although cachexia is a morbidity factor in human, there is to date no animal model of septic cachexia. The goal of this study is to create and characterize a murine model of septic cachexia and evaluate endogenous ghrelin variations. Among the factors involved in both sepsis and muscular protection, current research highlights a potential role for ghrelin in muscle protection but there are conflicting data regarding its variations  Introduction: Metabolism of tyrosine can be switched in conditions of hemodynamic instability and tissue hypoperfusion. The violation of the oxygen-dependent metabolism of tyrosine must be accompanied by the activation of alternative pathways ( Figure 1). Previously, we found a high level of content in the blood of aromatic intermediates p-hydroxyphenyllactic acid (p-HPLA) and p-hydroxyphenylacetic acid (p-HPAA) in patients with sepsis [1,2]. We assume that the anaerobic microbiota can take an important part in biodegradation of excess alternative metabolites of aromatic amino acids [3,4].   Introduction: Lipidose (formerly GR270773) is a protein-free phospholipid emulsion intended for the treatment of hospitalized patients with suspected or confirmed Gram-negative severe sepsis. Lipidose contains phosphatidylcholine, triglyceride and sodium cholate formulated to optimize delivery of the phospholipid component to the surface of high-density lipoprotein (HDL) and other lipoproteins, thereby enhancing the capacity of the patient's circulating lipoprotein pool to bind and neutralize microbial toxins. When Lipidose is infused into blood, the cholic acid is adsorbed onto serum albumin and the phospholipid selectively associates with lipoproteins. Bound and neutralized toxins are removed from the circulation by the liver and excreted along with the cholic acid into the bile. The LIPOS trial enrolled 1,400+ patients at 235 study centers in 31 countries to access Lipidose treatment at two dose levels. The LIPOS headline data presented only a small mortality benefit for the lower dose and no benefit from the higher dose [1]. A subgroup analysis was carried out to test the hypothesis of benefit in the subgroup with adequate liver function, using serum albumin levels as a measure of liver function, and adequate pre-existing HDL or total lipoprotein to accept phospholipid as predicted by the mechanism of action. Methods: Albumin, cholesterol and HDL were measured in stored serum samples. The response to treatment and interactions with baseline covariates specified in LIPOS were tested after exclusion of subjects in the lowest biomarker quartiles (AlbTC25 and AlbHDL25). Results: Subjects above the lowest quartile of albumin cleared Lipidose significantly faster than those in the lowest quartile (P < 0.003). Interactions between treatment and planned use of intravenous stress replacement doses of corticosteroids (IVCST) were found in the AlbTC25 and AlbHDL25 subgroups (P < 0.05). Exclusion of these subjects revealed strong relationships between treatment benefit and cholesterol or HDL that were used to select optimal biomarker thresholds. Requiring albumin ≥1.5 g/dl and either cholesterol ≥1 mM or HDL ≥0.5 mM selected 59% and 36%, respectively, of the LIPOS population. Treatment with Lipidose reduced mortality in these subgroups by 6.6% (P < 0.025) or 10.8% (P < 0.005) respectively. The treatment benefits persisted for at least 1 year. Conclusion: A strong negative interaction with IVCST may have masked a significant treatment benefit in LIPOS. This interaction may be related to the ability of bile acids to slow clearance and raise concentrations of corticosteroids [2,3]. Biomarkers can be used to select subjects with early severe septic shock responsive to treatment with Lipidose.
Center, University of Amsterdam, the Netherlands; 5  Introduction: Community-acquired pneumonia (CAP) is the most common infectious reason for admission to ICUs and has mortality of up to 37% [1]. Highest mortality rates are in Gram-negative infections with lower rates in Streptococcus pneumoniae and viral infections [2]. Microbiology is difficult to establish with most prospective studies identifying agents in only 50% of cases [3]. We analysed microbial aetiology of CAP in ICU over 1 year and assessed its effects on inpatient mortality, length of mechanical ventilation and length of ICU stay. Methods: We retrospectively reviewed admissions to AMNCH ICU between February 2013 and February 2014 catalogued as having pneumonia from chest radiograph and clinical findings on the internal audit system (n = 91). After chart review, 28 were excluded as hospital-acquired pneumonias, 12 due to insufficient information and 21 due to primary diagnosis other than pneumonia. Thirty patients were selected on the basis that CAP was the likely reason for ICU admission.  and was associated with a shorter ventilation period (mean 2.66 days) and ICU stay (mean 6 days). Conclusion: Microbial aetiology was identified in a high proportion of patients (72%) admitted to ICU with CAP, reflecting timely collection of appropriate specimens. Infection with Gram-negative organisms had the highest mortality, length of mechanical ventilation and length of ICU stay, while the pathogens usually seen in CAP were associated with more favourable outcomes.
Introduction: Maintenance of the integrity of the endothelial barrier is crucial in pathological inflammatory/infectious conditions [1]. The endothelial barrier dysfunction leading to increased vascular permeability and leukocyte transmigration in the systemic inflammatory state has been intrinsically related to the multiple organ dysfunction syndrome. The Rho GTPases family regulates the organization of the actin cytoskeleton, and plays a fundamental role in maintaining homeostasis and function of the endothelial barrier [2]. The study of the mechanisms of intracellular signaling related to the integrity of the endothelial barrier may help in understanding the hemodynamic changes during systemic infection. Thus, this study aimed to correlate the pattern of genic expression of GTPases during the initial periods of sepsis in order to understand the kinetic of these molecular mechanisms in organs often affected in sepsis. Methods: Wistar rats weighting 200 to 250 g were submitted to: nonlethal sepsis (2 ml Escherichia coli 10 7 UFC/ml i.v. inoculation [3,4], n = 5, S7 group); minor trauma (cervical incision with catheter implantation in jugular, injection of 2 ml saline, n = 5, SHAM group); without any procedure (n = 5, NAIVE group). After 2 and 6 hours, the liver and kidney were collected to determine RhoA, Rac1 and Cdc42 gene expression by quantitative real-time PCR.
Results: Low expression of RhoA was observed in both organs (Figures 1A and Figure 2A). Rac1 and/or Cdc42 showed higher expression in both organs in animals submitted to systemic infection (S7 group) compared to the SHAM and NAÏVE groups in both periods (Figures 1B, C and Figure 2B, C). Although both organs showed a similar pattern, the kidney showed a statistically significant increase of Cdc42 and Rac1 as compared to the liver, showing that GTPase expression might differ, possibly due to endothelial heterogeneity of each organ to perform its specific function. These results suggest that the infectious process leads to a higher gene expression of Rac1 and/or Cdc42 in relation to a minor surgical trauma inflammation. The RhoA and Rac1 gene expression were inversely correlated, as reported in the literature [5].  (B) and Cdc42 (C) in the liver of rats. Animals were subjected to nonlethal sepsis 2 ml Escherichia coli 10 7 UFC/ml i.v. inoculation (S7 group), minor trauma 2 ml saline i.v. inoculation (SHAM group), or animals without inoculation were used as control group (NAÏVE group). After 2 or 6 hours the kidney was isolated. Total RNA was extracted and transcribed to cDNA. Gene expression was quantified by real-time PCR. Graph bars show the relative concentration of each cDNA compared with the NAIVE group and the housekeeping gene HPRT. Results are mean ± SEM (n = 5 animals/group) and analyzed in triplicate. *P < 0.05 compared to NAÏVE group Introduction: Throughout the world, the number of patients at risk for bloodstream infections (BSIs) continues to rise. BSIs are associated with high rates of morbidity and mortality, and they markedly increase the costs of hospital care. Prompt identification of the causative agent(s) and rapid initiation of appropriate antimicrobial therapy are critical for reducing mortality, especially in patients with septic shock. Matrix-assisted laser desorption-ionization time of flight (MALDI-TOF) equipment is increasingly used in the microbiological laboratory. The goal of the present investigation was to apply mass spectrometry directly from positive blood cultures in order to detect and identify bacterial strains using a simplified homemade protocol. Methods: One-hundred and sixty-six positive blood cultures (Bactec FX; Becton Dickinson) were analyzed. Gram and conventional plates were used and the identification was done by Wider System (Soria Melguizo). In the MALDI-TOF protocol, positive blood cultures were processed as follow: 5 ml sample of positive broth was extracted and centrifuged at 1,000 rpm for 10 minutes in order to remove blood cells. Introduction: Sepsis is a complex and heterogeneous syndrome in which inflammatory and infection mechanisms are implicated. Thus, it is difficult to differentiate those mechanisms in mammals where inflammation is a highly complex biological process. In our laboratory, besides studying sepsis in human blood biological samples, we choose to investigate a murine model in ordered to describe a global transcriptome overview of critical events occurring in the blood, brain and lung during an induced non-infection inflammation. We decided to complete our view with another animal model where the inflammatory process is less complex and mainly achieved through innate immunity Drosophila melanogaster.
Methods: Female C57BL6/J mice received an intravenous oleic acid (OA) injection to induce a controlled inflammation. Lung, brain and peripheral blood mononuclear cell (PBMC) expression patterns were analyzed using an Agilent 60K cDNA mouse microarray. The enrichment of canonical pathways revealed marked changes in pathways involving the immune and inflammatory responses. The inflammatory process in D. melanogaster shares common mechanisms in innate immunity with mammals (Toll pathway/TLR). We decided to set up a model, called the double-hit model, where drosophila females aged from 7 to 10 days have been injured with a needle (inflammatory hit) before being infected with a needle that has been dipped in a bacterial solution of Pseudomonas entomophila (infectious hit). The next step will be to investigate the transcriptome in order to highlight the differences and have a global view of biological mechanisms involved; and also look for genetic factors involved in this experiment. The DGRP (Drosophila Genomic Reference Panel) project is a panel of 200 inbred fly lines that have been fully sequenced.
Results: Strikingly, in mice models, all significant pathways identified in the brain were also significant in the lung. The inflammatory responses oscillated between proinflammatory and anti-inflammatory response in both the lung and brain, the time course, however, being different in the two organs. In PBMC, we observed a significant response delay after OA injection and the pathways identified differed from those identified in the lung and brain. Our second objective will be to use this model of the expression to set up a tool to analyze precisely the effect of an infection in a second hit during the sterile inflammation activation/repression time courses. In D. melanogaster, we recorded their survival in comparison with flies that only received the infectious hit and observed a difference of~40%. Flies that have received a non-infectious double-hit showed a 10% decrease in survival rate. We will perform the double-hit experiment with all the lines of the DGRP and then through genome-wide association study. We expected to identify single nucleotide polymorphisms that are associated with resistance or susceptibility to the double-hit. Conclusion: In mice, we assessed gene expression profiles in mouse lung, brain and PBMC associated with a lung inflammation during a 24-hour time course. Overall, our microarray analysis provides a global and detailed overview of critical transcriptional events occurring in the blood, brain and lung. The analysis of gene functional annotation revealed several major features. First, many genes were upregulated or downregulated over the time in the lung, brain, and blood. Second, the analysis revealed pathways clearly related to inflammation mainly in the lung and brain. The strong inflammation observed in the brain, and a limited inflammation in the blood, indicates that blood gene expression profiles poorly reflect those observed in the lung and brain. Third, all the pathways identified in the brain were also identified in the lung, although a few common genes were characterized; it should also be stressed that there was a delay response in the brain.

P68
Role of nonpneumoniae mycoplasma in the pathogenesis of ventilator-associated pneumonia: an in vitro assessment.  (Figure 1) [2]. The aim of this study was to examine the effect of M. salivarium on human immune cells in vitro. Specifically, we measured cytokine production and phagocytosis activity in response to M. salivarium exposure. Methods: Whole human blood was obtained from healthy donor volunteers and cell types were isolated using diffusion gradients and magnetic labeling as appropriate. Monocytes and macrophages were incubated with M. salivarium for 24 hours before a subsequent LPS stimulus.
Macrophage phagocytosis assays were conducted after exposure times of 60 minutes and 24 hours to M. salivarium. Cytokines were measured using ELISA and human cytokine bead array kits.
Results: There was a statistically significant decrease in phagocytosis between control cells and the macrophages exposed to both a low titer of M. salivarium (P value 0.018) and a medium titer of M. salivarium (P value 0.011) after 24 hours of exposure ( Figure 2). There was a statistically significant decrease in phagocytosis activity between the macrophages exposed to the medium titer of M. salivarium for 24 hours versus 60 minutes (P value 0.013). Exposure of macrophages to mycoplasma resulted in decreased release of TNFα after a subsequent LPS stimulus (Figure 3). To our knowledge, this is the first time extracellular traps have been induced in macrophages in response to M. salivarium (Figure 4). Conclusion: Although further research is needed, it is interesting that the presence of M. salivarium caused an anti-inflammatory effect as well as impaired antigen presentation secondary to impaired phagocytosis. This could be consistent with the better outcome in mechanically ventilated patients that did not have M. salivarium bacteria detected in their bronchoalveolar lavage washings. Extracellular traps contribute to microbial containment by forming a physical barrier composed of chromatin and cytoplasmic proteins to enhance antimicrobial synergy while minimizing damage to host tissues [3]. It is interesting that M. salivarium induced extracellular traps.

P69
Probiotic pretreatment improves survival and prevents gut mucosal barrier dysfunction in sepsis. Introduction: The gut is the largest immune organ and plays a central role in the promotion of systemic inflammatory responses [1]. Perturbations of intestinal epithelial homeostasis during sepsis include increased proinflammatory cytokine production, increased intestinal permeability and apoptosis [2][3][4][5][6]. Healthy gut is essential to promote host health and prevent organ dysfunction in sepsis. Probiotics seem to keep gut homeostasis through different pathways, such as the modulation of microbial activity, energy regulation, anti-inflammatory cytokine production, gene expression  and cell differentiation [7]. Probiotics have been shown an effective treatment in various clinical conditions, although the potential benefits of probiotic treatment in sepsis remain largely undefined. The aim of the present study was to investigate the effect of probiotic treatment on gut dysfunction and inflammatory signaling in septic rats.
Methods: Sepsis was induced by cecal ligation and puncture (CLP) in Wistar male rats (8 weeks old). They were pretreated with probiotics or vehicle once a day during 7 days before CLP. The chosen probiotic mixture contained 10 × 10 7 CFU Bifidobacterium longum, 10 × 10 6 CFU Lactobacillus bulgaricus, 10 × 10 6 CFU Lactobacillus acidophilus. Colonic tissue and serum samples were collected 24 hours after CLP for ELISA and protein expression analysis by western blotting.
Results: Our data demonstrate that probiotic pretreatment improved survival of septic rats ( Figure 1) and this effect is accompanied by a marked decrease of IL-1β and TNFα (Figure 2A,B). Sepsis leads to severe intestinal epithelial damage with a decrease in claudin 2 and occludin protein expression ( Figure 3A,B); probiotic pretreatment reversed these alterations in parallel with an increase in Hsp72 and Hsp25 activation ( Figure 4A,B). In intestinal epithelial cells, the inducible Hsp have been shown to preserve tight junction and barrier function. The maintenance of epithelial barrier integrity induced by probiotic pretreatment, in parallel with an activation of cytoprotective pathway, may culminate in the restoration of the intestinal epithelial function. Conclusion: Our results show that probiotics pretreatment fulfills a dual function at the intestinal mucosa: in addition to preventing intestinal permeability disruption, it also attenuates proinflammatory cytokine release, diminishing the exacerbate host's reaction to infection and offering a novel prophylactic strategy to sepsis.
Introduction: The amount of fluid during resuscitation of septic shock is important. Too little fluid may result in tissue hypoperfusion; however, too much fluid may result in volume accumulation. Several recent studies have demonstrated that a positive fluid balance in critical illness is associated with deteriorating outcomes. However, some studies have shown opposite results. The objective of this study was to determine whether initial fluid balance in septic shock patients is correlated with ICU mortality.
Methods: This is a retrospective study of septic shock patients admitted to a mixed medical-coronary care unit of Songklanagarind hospital from 2005 to 2011. Multivariate logistic regression analysis was used to identify predictors of mortality.
Results: A total of 1,048 patients admitted to ICU for septic shock was divided into two groups: in-ICU survivors (n = 555 (53%)) and nonsurvivors (n = 493 (47%)). Median survival time was 10 days (95% CI: 8 to 12 days). The respiratory tract was the most common site of infection (47.6%). Community-acquired infections accounted for 59.6%. Survivors were older than nonsurvivors (62 vs. 56 years, P = 0.016). Nonsurvivors were more severely ill and had shorter ICU stays (2 vs. 5 days, P < 0.001). Nonsurvivors received albumin and steroid more than survivors. Median cumulative fluid at 24, 48 and 72 hours of septic shock onset were 4.2, 7.7 and 10.5 l respectively. Nonsurvivors had significantly larger median cumulative fluid intake at 24 hours (4.6 vs. 3.9 l, P < 0.001), at 48 hours (8.2 vs. 7.1 l, P < 0.001) and at 72 hours (11.4 vs. 9.9 l, P < 0.001). Nonsurvivors also had In multivariate logistic regression analysis, factors significantly associated with ICU mortality were mean fluid balance, APACHE II score, SOFA score, length of ICU stay, ARDS, steroid use, parenteral nutrition use and source of infection. Conclusion: A more positive cumulatively fluid balance over 3 days is associated with ICU mortality in septic shock. Multivariate analysis found not only nonmodifiable factors such as severity score, source of infection, length of ICU stay and ARDS, but also modifiable factors such as parenteral nutrition use, steroid use and mean fluid balance were significantly associated with mortality.

P71
Alpha lipoic acid attenuates oxidative stress-induced damage macromolecules in the brain of rats with sepsis-associated encephalopathy.
LG Introduction: Pathophysiological mechanisms of sepsis-associated encephalopathy involve oxidative stress. This imbalance between the prooxidant and antioxidant causes damage to macromolecules such as lipids and proteins, thus the employment of antioxidants becomes an attractive proposition. Alpha lipoic acid (LA), a potent antioxidant, is able to cross the blood-brain barrier, and is an important cofactor in enzymatic and cellular energy metabolism. We aimed to determine the use of AL in oxidative damage and neutrophil infiltration in rat brain 12 and 24 hours after induction of sepsis model by cecal ligation and puncture (CLP). Methods: Male Wistar rats (250 to 350 g) were subjected to CLP model, with sham control. Groups were divided into sham + saline, sham + AL, CLP + saline and CLP + AL (200 mg/kg orally with single administration after CLP), n = 10. At 12 and 24 hours, rats were euthanized, the hippocampus, striatum, cerebellum, cortex and prefrontal cortex removed, lipid peroxidation assessed by TBARS, damage to proteins by protein carbonylation, myeloperoxidase activity (MPO) and the formation of nitrite and nitrate. Data were analyzed by ANOVA with post hoc Tukey test and log-rank test with P < 0.05. Results: In 12 hours compared with the CLP group, the CLP + AL group showed a reduction in lipid peroxidation in the striatum, in the protein carbonylation in the cortex and hippocampus, in the MPO activity in the striatum and hippocampus, and decreased formation of nitrite and nitrate in the hippocampus and cortex. Conclusion: While differences were not observed in 24 hours, TBARS found protein carbonylation in a reduction of damage to the CLP + AL group over the cerebellum, MPO in the striatum, hippocampus and prefrontal, and hippocampus, cerebellum, and prefrontal to nitrite/nitrate. AL may be an important therapeutic target in reducing neurologic complications in animal models of sepsis. Acknowledgements: Financial support from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Universidade do Sul de Santa Catarina (UNISUL)  Introduction: Septic shock is a medical emergency that, despite the medical advances that have been made, still remains a major cause of hospital deaths. Cell therapy is an innovative field of research that could provide a therapy for sepsis. Mesenchymal stromal cells (MSC) are promising in cell therapy and more importantly for sepsis because of their immunosuppressive capabilities [1]. MSC have been shown by several groups to have a positive effect against sepsis in vivo [2][3][4]. It has been predicted that the MSC interact with macrophage to release IL-10 that in turns reduces inflammation [4]. Other groups have focused on the use of stimulated MSC to ameliorate their immunosuppressive capabilities [5]. The main stimulation of MSC has been the use of inflammatory stimulants like IFNγ. Our work focuses on the identification of effective MSC donors, whether primed with IFNγ or naïve, and the development of in vitro models that will predict how an MSC donor will act in vivo. We also want to eliminate the use of cells completely and use their secreted microvesicles as a therapy. The hypothesis is that the in vitro models will eliminate a noneffective MSC donor and allow us to identify the MSC donor that will have the greatest effect. Methods: We developed two in vitro models that are similar to what happens in vivo with WBC as they circulate in a septic patient. The first test is the adherence of WBC to a layer of HUVECs in the presence of MSC or microvesicles. The second is a permeability test to determine MSC ability to block the permeability of a HUVEC layer. Results: Our preliminary results have shown that we are able to identify, using our two in vitro models, which MSC donor would be an effective MSC for cell therapy.
Conclusion: MSC and their paracrine factors have to the potential to be an effective therapy for sepsis, but one needs to identify an effective donor before use in cell therapy.
value more than 0.55 is a meaningful value for early diagnosis of sepsis due to Gram-negative infection.

P75
Effect of estimated glomerular filtration rate and fluid balance on clinical course and outcomes of children admitted with severe dengue. Introduction: Dengue fever is one of the most important seasonal epidemics in Asia Case fatality rates vary from 1 to 5% [1]. Mortality from severe dengue may range from 26% in DHF to as high as 47% in DSS [2,3]. The pathogenesis of shock in dengue fever (DF) is centered on increased capillary permeability in the critical phase leading to hypovolemia and shock in severe dengue. There have been multiple studies that compare fluid regimens in the management of dengue [4,5]. These studies do not assess the child's renal function and ability to handle the fluid loads. GFR <60 ml/ minute indicates a significant decrease in the renal functioning and there are no pediatric studies that examine their association with in-hospital stay and outcomes in children with severe dengue. With this introduction we formulated this study protocol to examine that association. The objectives were to measure the estimated glomerular filtration rate (eGFR) at admission and fluid balance in the first 36 hours of ICU stay and assess their effect on disease course and outcomes in severe dengue.
Methods: This was designed as a retrospective descriptive study in a tertiary-level pediatric ICU in South India. Case records of all children fulfilling the WHO case definition of severe dengue were included, those who received intravenous fluid for less than 12 hours were excluded. Primary parameters measured included fluid balance in the first 36 hours measured every 12 hours, durations of oxygen requirement, mechanical ventilation, ICU stay and total hospital stay. Outcomes measured were death and survival. Results: Twenty-six children were enrolled, 14 boys and 12 girls. The median duration of ICU stay was 60 hours, and that of hospital stay 109 hours. eGFR was less than 60 ml/minute in six patients (83.3% expired and 16.7% survived). eGFR, measured by modified Schwartz's formula, at the time of admission correlated inversely with requirement of oxygen therapy and mechanical ventilation (P < 0.05) and fluid balance in the first 36 hours. Positive fluid balance (FO > 15%) in the first 36 hours was significantly higher in children who expired (P = 0.011). eGFR <90 ml/minute at admission had 100% sensitivity and 79% specificity to predict the possible occurrence of fluid overload >15% (area under curve = 0.882). Conclusion: Fluid balance in the first 36 hours had a significant positive correlation with mortality and negative correlation with eGFR. Children with admission eGFR <90 ml/minute may require restrictive fluid therapy to improve survival.
Introduction: This audit evaluates the adherence to the prophylactic surgical antibiotic policy at Kingston Hospital. With a greater understanding of the current use of prophylactic surgical antibiotics comes the ability to improve patient care while minimising the development of antibiotic resistance. Surgical site infections are a major source of hospital-acquired infections, causing significant morbidity and mortality. In appropriate cases, surgical antibiotic prophylaxis is essential in preventing such infections; however, this comes with increased risks of antibiotic resistance and antibiotic-associated diarrhoea. Consequently, this institution has extensive guidelines as to the cases in which antibiotic prophylaxis is required and what antibiotics should be administered. This audit examines the adherence to these guidelines. We audited against our local antibiotic prescribing policy, named the Blue Book guidelines. Methods: Notes were audited retrospectively for 80 patients undergoing surgery between 19 August and 18 December 2013. These were audited against local antibiotic guidelines (Blue Book guidelines) and analysed using descriptive statistics. Results: Only 57% of operations were compliant with Blue Book antibiotic prophylaxis guidelines. For operations where administration of antibiotics was appropriate, 24% of patients received the incorrect choice, 19% were given the incorrect dose, 28% were given antibiotics for an inappropriate time relative to the procedure, and 38% received antibiotics for the incorrect duration. For operations not requiring prophylaxis, 48% of patients incorrectly received antibiotics. Conclusion: Compliance to local guidelines for prophylactic antibiotics is extremely poor. This not only risks increased morbidity and mortality from surgical site infections but also risks the development of antibiotic resistance and Clostridium difficile colitis. We suggest enhancement of the antibiotic guidelines, and education of those using them. This will improve the provision of antibiotic prophylaxis in the Trust, and ultimately improve patient care.

P77
Presepsis biomarker: high-density lipoprotein. Introduction: Delay in diagnosis and initiation of antibiotic treatment has been shown to increase mortality. Biomarkers can play an important role in diagnosis and prognosis of sepsis. We aimed to evaluate the correlation between septicemia and high-density lipoprotein (HDL) level in burned patients. Methods: A prospective study conducted at Al-Sadr teaching hospital, Maysan, Iraq, during the period from April to September 2013. Blood samples were collected from patients every other day to measure the level of HDL and triglycerides. Other blood samples were collected in blood culture tubes for culturing to verify septicemia depending on the clinical evidence. Results: Seventy-five patients (Table 1) were admitted consecutively into the burn unit, 35 of them (46%) developed septicemia and 11 of the 35 patients died. All dead patients had HDL value <5 mg/dl 1 or 2 days before dying since our blood samples were collected every 2 days (Tables 2 and 3). Other laboratory tests such as WBCs, platelet account, albumin level, and so forth were made to confirm sepsis (Table 4). A comparison between the level of lipid profile before and after sepsis showed a significant drop in HDL level during the onset of sepsis (Table 5). We also found that patients with HDL value <15 mg/dl were at high risk of developing sepsis.
Conclusion: There was a strong correlation between HDL level and septicemia in burn patients. The HDL value is a good biomarker for sepsis; it decreases below normal level and continues to diminish and reach an immeasurable level at the advanced stage of septicemia.