Share
array:23 [ "pii" => "S217357272100151X" "issn" => "21735727" "doi" => "10.1016/j.medine.2020.05.013" "estado" => "S300" "fechaPublicacion" => "2021-12-01" "aid" => "1524" "copyright" => "Elsevier España, S.L.U. and SEMICYUC" "copyrightAnyo" => "2020" "documento" => "article" "crossmark" => 1 "subdocumento" => "fla" "cita" => "Med Intensiva. 2021;45:541-51" "abierto" => array:3 [ "ES" => true "ES2" => true "LATM" => true ] "gratuito" => true "lecturas" => array:1 [ "total" => 0 ] "itemSiguiente" => array:18 [ "pii" => "S2173572721001557" "issn" => "21735727" "doi" => "10.1016/j.medine.2020.10.006" "estado" => "S300" "fechaPublicacion" => "2021-12-01" "aid" => "1603" "copyright" => "Elsevier España, S.L.U. and SEMICYUC" "documento" => "article" "crossmark" => 1 "subdocumento" => "rev" "cita" => "Med Intensiva. 2021;45:552-62" "abierto" => array:3 [ "ES" => true "ES2" => true "LATM" => true ] "gratuito" => true "lecturas" => array:1 [ "total" => 0 ] "en" => array:13 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Review</span>" "titulo" => "Ten answers to key questions for fluid management in intensive care" "tienePdf" => "en" "tieneTextoCompleto" => "en" "tieneResumen" => array:2 [ 0 => "en" 1 => "es" ] "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "552" "paginaFinal" => "562" ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "Diez respuestas a preguntas clave para el manejo de fluidos en cuidados intensivos" ] ] "contieneResumen" => array:2 [ "en" => true "es" => true ] "contieneTextoCompleto" => array:1 [ "en" => true ] "contienePdf" => array:1 [ "en" => true ] "resumenGrafico" => array:2 [ "original" => 0 "multimedia" => array:7 [ "identificador" => "fig0005" "etiqueta" => "Figure 1" "tipo" => "MULTIMEDIAFIGURA" "mostrarFloat" => true "mostrarDisplay" => false "figura" => array:1 [ 0 => array:4 [ "imagen" => "gr1.jpeg" "Alto" => 1566 "Ancho" => 2175 "Tamanyo" => 249909 ] ] "descripcion" => array:1 [ "en" => "<p id="spar0025" class="elsevierStyleSimplePara elsevierViewall">Factors conditioning the oxygen delivery (DO<span class="elsevierStyleInf">2</span>) to tissues and interventions from hemodynamic perspectives. The right side of the figure is intentionally left in shadow as the focus of the review is not the approach from transfusion and ventilatory perspective. CaO<span class="elsevierStyleInf">2</span>: arterial concentration of oxygen; Hb: hemoglobin; SaO<span class="elsevierStyleInf">2</span>: saturation of oxygen.</p>" ] ] ] "autores" => array:1 [ 0 => array:2 [ "autoresLista" => "F. Sanfilippo, A. Messina, M. Cecconi, M. Astuto" "autores" => array:4 [ 0 => array:2 [ "nombre" => "F." "apellidos" => "Sanfilippo" ] 1 => array:2 [ "nombre" => "A." "apellidos" => "Messina" ] 2 => array:2 [ "nombre" => "M." "apellidos" => "Cecconi" ] 3 => array:2 [ "nombre" => "M." "apellidos" => "Astuto" ] ] ] ] ] "idiomaDefecto" => "en" "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2173572721001557?idApp=WMIE" "url" => "/21735727/0000004500000009/v2_202201010839/S2173572721001557/v2_202201010839/en/main.assets" ] "itemAnterior" => array:18 [ "pii" => "S2173572721001521" "issn" => "21735727" "doi" => "10.1016/j.medine.2020.05.014" "estado" => "S300" "fechaPublicacion" => "2021-12-01" "aid" => "1527" "copyright" => "Elsevier España, S.L.U. and SEMICYUC" "documento" => "article" "crossmark" => 1 "subdocumento" => "fla" "cita" => "Med Intensiva. 2021;45:532-40" "abierto" => array:3 [ "ES" => true "ES2" => true "LATM" => true ] "gratuito" => true "lecturas" => array:1 [ "total" => 0 ] "en" => array:13 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Original</span>" "titulo" => "Hemodynamic resuscitation with fluids bolus and norepinephrine increases severity of lung damage in an experimental model of septic shock" "tienePdf" => "en" "tieneTextoCompleto" => "en" "tieneResumen" => array:2 [ 0 => "en" 1 => "es" ] "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "532" "paginaFinal" => "540" ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "La resucitación hemodinámica con bolos de fluidos y noradrenalina incrementa la severidad del daño pulmonar en un modelo experimental de <span class="elsevierStyleItalic">shock</span> séptico" ] ] "contieneResumen" => array:2 [ "en" => true "es" => true ] "contieneTextoCompleto" => array:1 [ "en" => true ] "contienePdf" => array:1 [ "en" => true ] "resumenGrafico" => array:2 [ "original" => 0 "multimedia" => array:7 [ "identificador" => "fig0010" "etiqueta" => "Figure 2" "tipo" => "MULTIMEDIAFIGURA" "mostrarFloat" => true "mostrarDisplay" => false "figura" => array:1 [ 0 => array:4 [ "imagen" => "gr2.jpeg" "Alto" => 1542 "Ancho" => 2508 "Tamanyo" => 415154 ] ] "descripcion" => array:1 [ "en" => "<p id="spar0090" class="elsevierStyleSimplePara elsevierViewall">Evolution over time of hemodynamic measurements of each group. Circles represent mean values and vertical lines are SD. SHAM: Sham-operated group, ETX-NR: Non-resuscitated septic group. ETX-R: Resuscitated septic group. Cardiac index (CI), mean arterial pressure (MAP), stroke volume (SV) and heart rate (HR). When ETX was administered there was an increase in the CI (secondary to an increase in HR), with a progressive reduction in MAP. The administration of fluids increased CI, SV and MAP; However MAP's baseline levels were not reached, so all animals required dose of norepinephrine. *<span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.005 ETX-R vs. basal. <elsevierMultimedia ident="202201010839440881"></elsevierMultimedia><span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.005 ETX-R vs. SHAM. #<span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.005 ETX-R vs. ETX-NR.</p>" ] ] ] "autores" => array:1 [ 0 => array:2 [ "autoresLista" => "P. Guijo Gonzalez, M.A. Gracia Romero, A. Gil Cano, M. Garcia Rojo, M. Cecconi, I.M. Monge Garcia" "autores" => array:6 [ 0 => array:2 [ "nombre" => "P." "apellidos" => "Guijo Gonzalez" ] 1 => array:2 [ "nombre" => "M.A." "apellidos" => "Gracia Romero" ] 2 => array:2 [ "nombre" => "A." "apellidos" => "Gil Cano" ] 3 => array:2 [ "nombre" => "M." "apellidos" => "Garcia Rojo" ] 4 => array:2 [ "nombre" => "M." "apellidos" => "Cecconi" ] 5 => array:2 [ "nombre" => "I.M." "apellidos" => "Monge Garcia" ] ] ] ] ] "idiomaDefecto" => "en" "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2173572721001521?idApp=WMIE" "url" => "/21735727/0000004500000009/v2_202201010839/S2173572721001521/v2_202201010839/en/main.assets" ] "en" => array:20 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Original</span>" "titulo" => "Clinical risk factors for early mortality in patients with community-acquired septic shock. The importance of adequate source control" "tieneTextoCompleto" => true "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "541" "paginaFinal" => "551" ] ] "autores" => array:1 [ 0 => array:4 [ "autoresLista" => "J. Vallés, E. Diaz, J. Carles Oliva, M. Martínez, A. Navas, J. Mesquida, G. Gruartmoner, C. de Haro, J. Mestre, C. Guía, A. Rodriguez, A. Ochagavía" "autores" => array:12 [ 0 => array:4 [ "nombre" => "J." "apellidos" => "Vallés" "email" => array:1 [ 0 => "jvalles1953@hotmail.com" ] "referencia" => array:1 [ 0 => array:2 [ "etiqueta" => "<span class="elsevierStyleSup">*</span>" "identificador" => "cor0005" ] ] ] 1 => array:2 [ "nombre" => "E." "apellidos" => "Diaz" ] 2 => array:2 [ "nombre" => "J." "apellidos" => "Carles Oliva" ] 3 => array:2 [ "nombre" => "M." "apellidos" => "Martínez" ] 4 => array:2 [ "nombre" => "A." "apellidos" => "Navas" ] 5 => array:2 [ "nombre" => "J." "apellidos" => "Mesquida" ] 6 => array:2 [ "nombre" => "G." "apellidos" => "Gruartmoner" ] 7 => array:2 [ "nombre" => "C." "apellidos" => "de Haro" ] 8 => array:2 [ "nombre" => "J." "apellidos" => "Mestre" ] 9 => array:2 [ "nombre" => "C." "apellidos" => "Guía" ] 10 => array:2 [ "nombre" => "A." "apellidos" => "Rodriguez" ] 11 => array:2 [ "nombre" => "A." "apellidos" => "Ochagavía" ] ] "afiliaciones" => array:1 [ 0 => array:2 [ "entidad" => "Critical Care Department, Fundació Parc Taulí, Hospital Universitari Parc Taulí, Sabadell, Spain" "identificador" => "aff0005" ] ] "correspondencia" => array:1 [ 0 => array:3 [ "identificador" => "cor0005" "etiqueta" => "⁎" "correspondencia" => "Corresponding author." ] ] ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "Factores de riesgo clínico asociados a mortalidad precoz en pacientes con <span class="elsevierStyleItalic">shock</span> séptico de origen comunitario. La importancia de un control del foco adecuado" ] ] "resumenGrafico" => array:2 [ "original" => 0 "multimedia" => array:7 [ "identificador" => "fig0005" "etiqueta" => "Figure 1" "tipo" => "MULTIMEDIAFIGURA" "mostrarFloat" => true "mostrarDisplay" => false "figura" => array:1 [ 0 => array:4 [ "imagen" => "gr1.jpeg" "Alto" => 1222 "Ancho" => 2084 "Tamanyo" => 106645 ] ] "descripcion" => array:1 [ "en" => "<p id="spar0085" class="elsevierStyleSimplePara elsevierViewall">Distribution of annual early and late deaths during the period of study (<span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span>=<span class="elsevierStyleHsp" style=""></span>0.576).</p>" ] ] ] "textoCompleto" => "<span class="elsevierStyleSections"><span id="sec0005" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0105">Introduction</span><p id="par0005" class="elsevierStylePara elsevierViewall">Septic shock is the most severe form of sepsis and occurs as a result of both community-acquired and health care-associated infections. Pneumonia, followed by intraabdominal and urinary tract infections are the most common causes.<a class="elsevierStyleCrossRefs" href="#bib0180"><span class="elsevierStyleSup">1–4</span></a></p><p id="par0010" class="elsevierStylePara elsevierViewall">Despite major technological advances, mortality of septic shock remains high, ranging from 20 to 50%.<a class="elsevierStyleCrossRefs" href="#bib0200"><span class="elsevierStyleSup">5–7</span></a> However, recent reports have shown decreases in mortality, associated to advances in training, better surveillance and monitoring, and prompt initiation of therapy to treat the underlying infection and support failing organs.<a class="elsevierStyleCrossRefs" href="#bib0215"><span class="elsevierStyleSup">8–12</span></a></p><p id="par0015" class="elsevierStylePara elsevierViewall">Most studies in patients with septic shock have analyzed ICU and/or hospital mortality, but few studies have evaluated the time in which these patients die.<a class="elsevierStyleCrossRefs" href="#bib0240"><span class="elsevierStyleSup">13–15</span></a> The most recent study to analyze the time of death of patients with septic shock found that 32% of patients with septic shock died within 72<span class="elsevierStyleHsp" style=""></span>h of admission to the ICU, and that these early deaths were mainly attributable to intractable multiple organ failure.<a class="elsevierStyleCrossRef" href="#bib0240"><span class="elsevierStyleSup">13</span></a></p><p id="par0020" class="elsevierStylePara elsevierViewall">The main objective of the present study was to estimate the annual incidence rates of mortality within 72<span class="elsevierStyleHsp" style=""></span>h after ICU admission of community-acquired septic shock (CASS) through a secondary analysis of recently published data from a 14-year period.<a class="elsevierStyleCrossRef" href="#bib0235"><span class="elsevierStyleSup">12</span></a> Secondary objectives included characterizing these patients and identifying risk factors associated with early mortality (EM) with the aim of improving survival in these patients.</p></span><span id="sec0010" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0110">Material and methods</span><span id="sec0015" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0115">Study population</span><p id="par0025" class="elsevierStylePara elsevierViewall">This is a secondary analysis of a retrospective observational cohort of consecutive adult patients (aged<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>18 y) admitted to the ICU at a university hospital in Spain between January 1, 2003 and December 31, 2016 with a first-time diagnosis of CASS and no other obvious cause of shock.<a class="elsevierStyleCrossRef" href="#bib0235"><span class="elsevierStyleSup">12</span></a> The hospital's ethics committee approved the study and waived the need for informed consent due to the observational nature of the study (Ethics Committee Corporació Sanitaria Universitaria Parc Taulí, 2018/610).</p></span><span id="sec0020" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0120">Data abstraction and definitions</span><p id="par0030" class="elsevierStylePara elsevierViewall">Trained research physicians used a standardized, piloted data form to collect data and select patients from a prospectively registered data base of all patients admitted to the ICU with community-acquired infection during the study period. Two investigators (JV and ED) with experience in infectious diseases were responsible for maintaining the database and ensuring the application of homogeneous diagnostic criteria during the study period. Septic shock was defined according to the Sepsis-3 definitions as the presence of hypotension requiring vasopressor therapy to maintain mean blood pressure<span class="elsevierStyleHsp" style=""></span>≥<span class="elsevierStyleHsp" style=""></span>65<span class="elsevierStyleHsp" style=""></span>mmHg during<span class="elsevierStyleHsp" style=""></span>≥<span class="elsevierStyleHsp" style=""></span>48<span class="elsevierStyleHsp" style=""></span>h, SOFA score<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>2 at admission to the ICU, and serum lactate<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>2<span class="elsevierStyleHsp" style=""></span>mmol/L after adequate fluid resuscitation.<a class="elsevierStyleCrossRef" href="#bib0255"><span class="elsevierStyleSup">16</span></a> Patients diagnosed with sepsis but who did not meet the diagnostic requirements for septic shock according to the Sepsis-3 definitions were not included in the definitive database.</p><p id="par0035" class="elsevierStylePara elsevierViewall">Variables collected included year of admission for septic shock, patient demographics (age, sex), and baseline comorbidities (VIH/AIDS, cancer, cirrhosis, chronic heart failure (≥ NYHA-II), chronic obstructive pulmonary disease (COPD), chronic renal failure requiring dialysis, diabetes mellitus, alcohol abuse, and immunodepression), diagnosed according to criteria reported elsewhere.<a class="elsevierStyleCrossRefs" href="#bib0235"><span class="elsevierStyleSup">12,17</span></a> Pathogens isolated from the local site and/or blood cultures within 48<span class="elsevierStyleHsp" style=""></span>h of ICU admission were considered potential causes of septic shock and were classified as negative culture, gram-positive bacteria, gram-negative bacteria, anaerobes, fungi or other (Influenza, <span class="elsevierStyleItalic">Pneumocystis jirovecii, Mycobacterium tuberculosis</span>). We also recorded the presence of bacteremia and fungemia and the appropriateness of antimicrobial agents. Initial empirical antibiotic treatment was defined as appropriate if an antibiotic prescribed within 24<span class="elsevierStyleHsp" style=""></span>h of the first encounter with the patient matched the in vitro susceptibility of a pathogen deemed to be the likely cause of infection. For episodes of septic shock with negative culture infections, the appropriateness of antibiotic therapy was not analyzed. Treatment decisions, including antibiotic prescriptions, were taken by attending physicians following protocols derived from locally adapted guidelines.</p><p id="par0040" class="elsevierStylePara elsevierViewall">Based on clinical diagnosis and/or isolation of pathogens, the anatomic focus of infection considered the source of sepsis was classified as respiratory, intraabdominal (peritonitis, abscess, small-bowel obstruction, spontaneous bacterial peritonitis, <span class="elsevierStyleItalic">Clostridium difficile</span>-associated colitis, perforated viscus, enterocolitis or diverticulitis, ischemic bowel, pancreatitis, or other), biliary (cholecystitis or cholangitis), genitourinary (pyelonephritis, obstructive uropathy-associated urinary tract infection, or gynecologic infections), skin or soft-tissue (cellulitis, abscess, or necrotizing fasciitis), or other.</p><p id="par0045" class="elsevierStylePara elsevierViewall">In episodes where a source control was susceptible, attending physicians decided on the most appropriate type: noninvasive (percutaneous/endoscopic) or invasive (surgical/laparoscopic). We recorded the time from hospital admission to source control and divided patients into those who received source control ≤12<span class="elsevierStyleHsp" style=""></span>h, which is the upper limit of the time recommended in the latest guidelines from the Surviving Sepsis Campaign<a class="elsevierStyleCrossRef" href="#bib0265"><span class="elsevierStyleSup">18</span></a> and those who received source control<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>12<span class="elsevierStyleHsp" style=""></span>h following onset. Adequate source control was defined as having a percutaneous, endoscopic, laparoscopic, or open procedure to drain infected fluid collections, debride infected tissues, and control ongoing enteric or other drainage producing peritonitis and if patients improved after the intervention (disappearance of fever, normalization of white blood counts, improvement of acute phase reactants, and a decrease in doses of vasopressors required) and required no further interventions on the same source during the ICU stay. In cases where these variables could not be reviewed, the effectiveness of source control was not evaluated.</p><p id="par0050" class="elsevierStylePara elsevierViewall">We also recorded the use of adjunctive therapies such as mechanical ventilation and renal replacement techniques as well as complications detected after admission to ICU that could be related to the natural history and/or management of septic shock such as acute kidney injury (defined as an increase in serum creatinine<span class="elsevierStyleHsp" style=""></span>≥<span class="elsevierStyleHsp" style=""></span>0.3<span class="elsevierStyleHsp" style=""></span>mg/dL within 48<span class="elsevierStyleHsp" style=""></span>h<a class="elsevierStyleCrossRef" href="#bib0270"><span class="elsevierStyleSup">19</span></a>), thrombocytopenia (defined as platelets<span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>150,000/μL<a class="elsevierStyleCrossRef" href="#bib0275"><span class="elsevierStyleSup">20</span></a>), and acute respiratory distress syndrome (ARDS) (defined as PaO<span class="elsevierStyleInf">2</span>/FiO<span class="elsevierStyleInf">2</span> <<span class="elsevierStyleHsp" style=""></span>300<span class="elsevierStyleHsp" style=""></span>mmHg with bilateral thoracic infiltrates in the absence of heart failure<a class="elsevierStyleCrossRef" href="#bib0280"><span class="elsevierStyleSup">21</span></a>).</p><p id="par0055" class="elsevierStylePara elsevierViewall">Patients’ severity at admission was recorded as APACHE II<a class="elsevierStyleCrossRef" href="#bib0285"><span class="elsevierStyleSup">22</span></a> score during the first 24<span class="elsevierStyleHsp" style=""></span>h after ICU admission and SOFA score<a class="elsevierStyleCrossRef" href="#bib0275"><span class="elsevierStyleSup">20</span></a> at admission to the ICU. Patients were followed up until death or discharge from the ICU and from the hospital.</p><p id="par0060" class="elsevierStylePara elsevierViewall">Based on the criteria used in a previous study in critically ill patients with septic shock,<a class="elsevierStyleCrossRef" href="#bib0240"><span class="elsevierStyleSup">13</span></a> we defined deaths occurring within 72<span class="elsevierStyleHsp" style=""></span>h after ICU admission as early and those occurring after this timepoint as late. Patients who received limited therapeutic effort during their stay in the ICU were excluded from the study.</p></span><span id="sec0025" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0125">Statistical analysis</span><p id="par0065" class="elsevierStylePara elsevierViewall">Continuous variables are expressed as means<span class="elsevierStyleHsp" style=""></span>±<span class="elsevierStyleHsp" style=""></span>standard deviations and categorical variables as frequencies and percentages. To compare the characteristics of patients who died early (≤3 days from ICU admission) with those of patients who survived<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>3 days, we used Student's <span class="elsevierStyleItalic">t</span>-tests for continuous variables and chi-square tests or Fisher's exact tests for categorical variables, as appropriate. Variables associated (<span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.1) with early ICU death in univariate analyses were included in a multivariate logistic models. The final logistic model obtained by a backward elimination procedure contained only statistically significant variables. Values of <span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.05 were considered statistically significant. Commercially available software (IBM® SPSS® Statistics for Windows, Version 25.0. Arkmon, NY) was used for all analyses.</p></span></span><span id="sec0030" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0130">Results</span><p id="par0070" class="elsevierStylePara elsevierViewall">During the study period, 625 patients were admitted to the ICU with CASS. The most common comorbidities were diabetes (29.1%) and cancer (20%), and 24% were immunocompromised. Infection was documented by a positive culture in 92.2% of patients. Blood cultures were positive in 52.5% of the patients. The most common site of infection causing sepsis was the lung (27%). The most frequent organ dysfunction associated with septic shock was acute kidney injury, which occurred in 555 (88.8%) patients, 132 (23.7%) of whom required continuous renal replacement. Thrombocytopenia was present in 390 (62.5%) patients. A total of 410 (65.7%) patients required invasive mechanical ventilation; ARDS was diagnosed in 142 (22.8%) patients.</p><p id="par0075" class="elsevierStylePara elsevierViewall">During the study period, the mean mortality related to septic shock was 25.8% in the ICU and 29.6% in the hospital. The mean EM related to septic shock was 14.4%. The percentages of early and late septic shock related deaths did not vary significantly over the years (<a class="elsevierStyleCrossRef" href="#fig0005">Fig. 1</a>). There was a trend toward decreased rates of both early and late mortality over the study period (<a class="elsevierStyleCrossRef" href="#fig0010">Fig. 2</a>).</p><elsevierMultimedia ident="fig0005"></elsevierMultimedia><elsevierMultimedia ident="fig0010"></elsevierMultimedia><p id="par0080" class="elsevierStylePara elsevierViewall">A total of 161 patients died in the ICU, 90 (55.9%) within the first 72<span class="elsevierStyleHsp" style=""></span>h of ICU admission; all these deaths were attributed to intractable multiple organ failure related to the primary infection. Of the 90 patients who died within the first 72<span class="elsevierStyleHsp" style=""></span>h, 49 (54.4%) died in the first 24<span class="elsevierStyleHsp" style=""></span>h, 21 (23.3%) in the second 24<span class="elsevierStyleHsp" style=""></span>h, and 20 (22.2%) in the third 24<span class="elsevierStyleHsp" style=""></span>h (<span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span><<span class="elsevierStyleHsp" style=""></span>0.001).</p><p id="par0085" class="elsevierStylePara elsevierViewall"><a class="elsevierStyleCrossRef" href="#tbl0005">Table 1</a> compares the characteristics of the patients who died in the first 72<span class="elsevierStyleHsp" style=""></span>h versus those of the patients who survived more than 72<span class="elsevierStyleHsp" style=""></span>h. Apache II and SOFA at admission were higher in patients who died in the first 72<span class="elsevierStyleHsp" style=""></span>h. Among comorbidities, only cirrhosis was significantly more frequent in the EM group. ARDS, thrombocytopenia, mechanical ventilation, and bacteremia were also more predominant in EM group.</p><elsevierMultimedia ident="tbl0005"></elsevierMultimedia><p id="par0090" class="elsevierStylePara elsevierViewall"><a class="elsevierStyleCrossRef" href="#tbl0010">Table 2</a> reports the anatomic sites of infection considered the source of sepsis and microbiology findings in the two groups. The sources of infection differed significantly between the groups and the proportion of patients who underwent source control was significantly lower in the EM group (41% vs. 53.2% in the group of patients that survived<span class="elsevierStyleHsp" style=""></span>><span class="elsevierStyleHsp" style=""></span>72<span class="elsevierStyleHsp" style=""></span>h, <span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span>=<span class="elsevierStyleHsp" style=""></span>0.019). Intraabdominal infections, skin and soft-tissue infections, and other sources that did not fit into the main predefined categories (mainly, endocarditis, meningitis, and unknown origin) were more common in the EM group, and the incidence of biliary and genitourinary infections was lower in this group. We observed no significant differences between the two groups in the pathogens responsible for the infections. There was no difference in the rate of appropriate antibiotic treatment between the two groups.</p><elsevierMultimedia ident="tbl0010"></elsevierMultimedia><p id="par0095" class="elsevierStylePara elsevierViewall">Source-control interventions were significantly less common in the EM group [37/90 (41.1%) patients vs. 285/585 (53.2%) patients who survived<span class="elsevierStyleHsp" style=""></span>≥<span class="elsevierStyleHsp" style=""></span>72<span class="elsevierStyleHsp" style=""></span>h, <span class="elsevierStyleItalic">p</span><span class="elsevierStyleHsp" style=""></span>=<span class="elsevierStyleHsp" sty