Free
Clinical Science  |   January 2008
Are Blood Transfusions Associated with Greater Mortality Rates?: Results of the Sepsis Occurrence in Acutely Ill Patients Study
Author Affiliations & Notes
  • Jean-Louis Vincent, M.D., Ph.D.
    *
  • Yasser Sakr, M.B., B.Ch., Ph.D.
  • Charles Sprung, M.D.
  • Svein Harboe, M.D.
    §
  • Pierre Damas, M.D.
  • * Professor, Department of Intensive Care, Erasme Hospital, Free University of Brussels. †Resident, Department of Anesthesiology and Intensive Care, Friedrich-Schiller-University, Jena, Germany. ‡Professor, Department of Anesthesiology and Critical Care Medicine, Hadassah Hebrew University Medical Center, Jerusalem, Israel. §Senior Physician, Department of Anesthesia, Division of Acute Care Medicine, Stavanger University Hospital, Stavanger, Norway. ∥Professor, Department of General Intensive Care, University Hospital Centre Sart-Tilman, Liege, Belgium.
Article Information
Clinical Science / Coagulation and Transfusion / Critical Care / Infectious Disease
Clinical Science   |   January 2008
Are Blood Transfusions Associated with Greater Mortality Rates?: Results of the Sepsis Occurrence in Acutely Ill Patients Study
Anesthesiology 1 2008, Vol.108, 31-39. doi:10.1097/01.anes.0000296070.75956.40
Anesthesiology 1 2008, Vol.108, 31-39. doi:10.1097/01.anes.0000296070.75956.40
ALTHOUGH blood transfusion can be life-saving in extreme circumstances, in the absence of life-threatening hemorrhage, the topic of transfusion is somewhat controversial. Blood transfusions have well-recognized problems, including the need to type and cross-match, and the potential transmission of diseases, such as hepatitis B and C, human immunodeficiency virus, and prions.1 Transfusion-induced immunosuppression is also a potentially important factor.2 However, anemia too has its own problems and is associated with increased morbidity and mortality.3,4 Determining who and when to transfuse is thus a challenge for intensivists, and recent years have seen continuing debate and discussion regarding the optimal transfusion “trigger.”
In an important Canadian trial, Hebert et al.  5 randomly assigned 838 critically ill patients to either a liberal protocol where transfusions were administered to maintain hemoglobin levels between 10 and 12 g/dl or a restricted strategy where hemoglobin levels were kept between 7 and 9 g/dl. Overall, the 30-day mortality rates were 19% in the restricted group and 23% in the liberal transfusion group (P  = not significant), and these differences were significant in younger (age <55 yr, 5.7 vs.  13%; P  = 0.02) and less sick (Acute Physiology and Chronic Health Evaluation II score ≤ 20, 8.7 vs.  16.1%; P  = 0.03) patients. The overall hospital mortality was significantly less in the restricted transfusion group (22 vs.  28%; P  = 0.05). These results had a definite influence on intensive care unit (ICU) practice, raising serious questions about the hitherto widely used 10 g/dl transfusion trigger, and encouraging intensivists to limit somewhat the use of transfusions. More recently, the Anemia and Blood Transfusion in Critical Care (ABC) study,6 an epidemiologic survey of 3,534 patients in 146 western European ICUs, confirmed increased mortality rates (ICU and hospital) in transfused patients. The increased mortality rates were maintained in a propensity analysis with patients matched for age, sex, disease severity, admitting hemoglobin level, recent history of hemorrhage or anemia, and hospital duration of stay. Here, the 28-day mortality rate was 22.7% in transfused patients and 17.1% in those who did not receive a transfusion (P  = 0.02). In a multivariate analysis, receipt of a blood transfusion increased the risk of dying by a factor of 1.4.6 Similar results were reported in a multicenter observational study of 4,892 patients in 284 ICUs across the United States, with the number of erythrocyte transfusions a patient received during the study being independently associated with longer ICU and hospital durations of stay and an increase in mortality.7 Studies in trauma patients,8 in patients with burns,9 in patients undergoing cardiac surgery,10 and in patients with acute coronary syndromes11 have also suggested increased mortality rates associated with transfusions in these groups of patients. Finally, in a recent study in pediatric critically ill patients, Lacroix et al.  12 reported that restricting transfusions to patients with a hemoglobin threshold of 7 g/dl was associated with decreased transfusion requirements and no increase in adverse events compared with patients transfused according to a trigger of 9.5 g/dl.
This is an important issue, and large, randomized controlled trials are necessary to determine what the current rules of play should be; in the meanwhile, epidemiologic studies can provide useful information on the current status of transfusion practice and on associated outcomes, provided they include multivariate analyses. The Sepsis Occurrence in Acutely Ill Patients (SOAP) study13 was designed to determine current ICU practice and the effects of that practice on outcomes for various topics, including blood transfusion.
Materials and Methods
Study Design
The SOAP study was a prospective, multicenter, observational study designed to evaluate the epidemiology of sepsis and other characteristics of ICU patients in European countries and was initiated by a working group of the European Society of Intensive Care Medicine (Brussels, Belgium). Institutional recruitment for participation was by open invitation from the study steering committee. Because this epidemiologic observational study did not require any deviation from routine medical practice, institutional review board approval was either waived or expedited in participating institutions, and informed consent was not required. We included all adult patients (older than 15 yr) admitted to the participating centers (see the 1for a list of participating countries and centers) between May 1 and May 15, 2002. Patients were followed up until death, until hospital discharge, or for 60 days. Those who stayed in the ICU for less than 24 h for routine postoperative observation were excluded.
Data Management
Data were collected prospectively using preprinted case report forms. Detailed instructions, explaining the aim of the study, instructions for data collection, and definitions for various important items were available for all participants on a dedicated Web site before starting data collection and throughout the study period. The steering committee processed all queries during data collection.
Data were entered centrally by medical personnel using the Statistical Package for Social Sciences (SPSS) version 11.0 for Windows (SPSS Inc., Chicago, IL). A random sample of 5% of the data were reentered by a different encoder and revised by a third one; a consistency of more than 99.5% per variable and 98.5% per patient was observed during the whole process of data entry. In case of inconsistency, data were verified and corrected. Daily frequency tables were revised for all variables, and the investigators were queried when data values were either questionable or missing for required fields. Data collection on admission included demographic data and comorbid diseases. Clinical and laboratory data for the Simplified Acute Physiology Score II14 were reported as the worst value within 24 h after admission. A daily evaluation of organ function that was based on a set of laboratory and clinical parameters according to the Sequential Organ Failure Assessment score15 was performed, with the most abnormal value for each of the six organ systems (respiratory, renal, cardiovascular, hepatic, coagulation, and neurologic) being collected on admission and every 24 h thereafter. For a single missing value, a replacement was calculated using the mean value of the results on either side of the absent result. When first or last values were missing, the nearest value was carried backward or forward, respectively. When more than one consecutive result was missing, it was considered to be a missing value in the analysis. Circulatory shock was defined as a cardiovascular Sequential Organ Failure Assessment score greater than 2, i.e.  , the use of dopamine at a dose greater than 5 μg · kg−1· min−1and/or epinephrine or norepinephrine at any dose. Sepsis, severe sepsis, and septic shock were defined according to the American College of Chest Physicians–Society of Critical Care Medicine consensus conference definitions.16 
Statistical Methods
Data were analyzed using SPSS version 11.0 for Windows and SAS version 9.1.3 software (SAS Institute Inc., Cary, NC). Descriptive statistics were computed for all study variables. The Kolmogorov–Smirnov test was used, and stratified distribution plots were examined to verify the normality of distribution of continuous variables. Nonparametric tests of comparison were used for variables evaluated as not normally distributed. Difference testing between groups was performed using the two-tailed t  test, Mann–Whitney U test, chi-square test, and Fisher exact test as appropriate. A Bonferroni correction was performed for multiple comparisons.
To determine the relative hazard of death due to blood transfusions, we performed a multivariate Cox proportional hazard model in the overall population with time to in-hospital death right censored at 30 days as the dependent factor, to avoid a positive effect of early mortality on duration of stay. Variables considered for the Cox regression analysis included age, sex, comorbid diseases, Simplified Acute Physiology Score II and Sequential Organ Failure Assessment score on admission, the type of admission (medical or surgical), the presence of sepsis during the ICU stay, and the country of origin. All variables were introduced in the model after exclusion of the presence of colinearity, based on R  2> 0.7. The time-dependent covariate method was used to check the proportional hazard assumption of the model; an extended Cox model was constructed, adding interaction terms that involve time, i.e.  , time-dependent variables, computed as the by-product of time and individual covariates in the model (time × covariate). Individual time-dependent covariates were introduced one by one and in combinations in the extended model, none of which were found to be significant (Wald chi-square statistics). To reduce the number of countries in the final model, the country effect was tested in a separate Cox regression model as a categorical variable with hospital mortality right censored at 30 days as the dependent variable. All countries that were not significant at P  < 0.2 were reduced to one category, and all of the countries were introduced in the final model. The need for blood transfusion was introduced in the model as a categorical variable.
Propensity scores17 were obtained through logistic regression of patient characteristics on blood transfusion status, i.e.  , need for blood transfusion as the dependent factor (table 1). The propensity score was calculated as the probability based on the final model. A greedy matching technique18 was used to match individual patients who received a blood transfusion at any time with individual patients who did not, based on propensity scores. The best-matched propensity score was identical to five digits. Once a match was made, the control patient was removed from the pool. This process was then repeated using four-digit matching, then three-digit matching, and so on. The process proceeded sequentially to a single-digit match on propensity score. If a match was not obtained at this point, the patient who had received a blood transfusion was excluded.
Table 1. The Logistic Regression Model Used to Calculate the Propensity Score 
Image Not Available
Table 1. The Logistic Regression Model Used to Calculate the Propensity Score 
×
The hazard of death was compared using an analysis of paired failure times19 with SAS Proc PHREG (SAS Institute Inc.) in the propensity score–matched pairs. To avoid survivor bias, an extended Cox proportional hazard model was constructed using the aforementioned covariates and adjusting for erythrocyte transfusion as a time-dependent covariate in the whole population and in the propensity-matched groups. All statistics were two-tailed, and P  < 0.05 was considered to be significant.
Results
Of 3,147 patients, 1,040 (33.0%) received a blood transfusion and 2,107 (67.0%) did not. Table 2presents the principal clinical data of the patients according to the transfusion status. Patients who received blood transfusions were somewhat older (mean age, 62 vs.  60 yr; P  = 0.035) and were more likely to have liver cirrhosis or hematologic cancer, to be a surgical admission, and to have an infection and sepsis. Transfused patients had a longer duration of ICU stay (5.9 vs.  2.5 days; P  < 0.001) and a higher ICU mortality rate (23.0 vs.  16.3%; P  < 0.001) but they were also more severely ill, as shown by a higher Simplified Acute Physiology Score II (40.2 vs.  34.7; P  < 0.001) and a higher Sequential Organ Failure Assessment score (6.5 vs.  4.5; P  < 0.001) on admission than the other patients (table 2). As anticipated, there was a direct relation between the number of blood transfusions and the mortality rate (table 3). Patients with longer ICU stays were more often transfused and received more units of blood (table 4). When asked whether they had been using leukodepleted blood at the time of the study, 76% of centers who replied stated that they had been routinely using leukodepleted blood, 19% stated that they had not been using leukodepleted blood, and 4% did not know.
Table 2. Characteristics of the Study Group 
Image Not Available
Table 2. Characteristics of the Study Group 
×
Table 3. Simplified Acute Physiology Score II, ICU and Hospital Mortality According to the Number of Transfused Units during ICU Stay 
Image Not Available
Table 3. Simplified Acute Physiology Score II, ICU and Hospital Mortality According to the Number of Transfused Units during ICU Stay 
×
Table 4. Transfusion Rate and Mean Number of Units Transfused According to ICU Duration of Stay 
Image Not Available
Table 4. Transfusion Rate and Mean Number of Units Transfused According to ICU Duration of Stay 
×
In a multivariate Cox regression analysis including sex and age, type of admission, main medical history (including cancer or hematologic cancer, cirrhosis, chronic lung disease), fluid balance, Simplified Acute Physiology Score II, and severity of organ dysfunction on admission as assessed by Sequential Organ Failure Assessment score, erythrocyte transfusion was not associated with an increased relative hazard of death at 30 days (relative hazard = 0.89; 95% confidence interval [CI]= 0.76–1.05; P  = 0.159; table 5).
Table 5. Multivariate Cox Proportional Hazard Analysis with Time to In-hospital Mortality at 30 Days as the Dependent Variable 
Image Not Available
Table 5. Multivariate Cox Proportional Hazard Analysis with Time to In-hospital Mortality at 30 Days as the Dependent Variable 
×
Eight hundred twenty-one pairs were matched according to the propensity score: 349 pairs (42.5%) at five digits, 342 pairs (41.7%) at four digits, and 130 pairs (15.8%) at three digits. ICU and hospital mortality rates were similar between transfused patients for whom matches could be identified and those who could not be matched (29 vs.  34.6%; P  = 0.134 and 22.1 vs.  26.5%; P  = 0.176, respectively). Baseline characteristics, severity scores, incidence of sepsis syndromes, and invasive procedures on the day of ICU admission were similar between the matched groups (table 6). There was a lower 30-day hazard of death in the transfused group compared with the other patients (hazard ratio = 0.73; 95% CI = 0.59–0.90; P  = 0.004; fig. 1).
Table 6. Patient Characteristics by Transfusion Status for the Propensity-matched Patients 
Image Not Available
Table 6. Patient Characteristics by Transfusion Status for the Propensity-matched Patients 
×
Fig. 1. Cumulative hazard of death in the propensity-matched groups (  solid line  = transfusion;  dashed line  = no transfusion). Hazard ratio = 0.73; 95% confidence interval = 0.59–0.90;  P  = 0.004. 
Image Not Available
Fig. 1. Cumulative hazard of death in the propensity-matched groups (  solid line  = transfusion;  dashed line  = no transfusion). Hazard ratio = 0.73; 95% confidence interval = 0.59–0.90;  P  = 0.004. 
×
In an extended Cox proportional hazard analysis, adjusting for erythrocyte transfusion as a time-dependent variable, erythrocyte transfusion was associated with a decreased relative hazard of death at 30 days in the whole population (relative hazard = 0.69; 95% CI = 0.48–1.01; P  = 0.055) and in the propensity score–matched groups (relative hazard = 0.57; 95% CI = 0.36–0.9; P  = 0.016).
Discussion
In this observational study, patients who received blood transfusions had higher ICU and hospital mortality rates than those who did not. This may be expected because blood transfusions are an index of severity of illness. However, after adjusting for confounding factors using a Cox regression model or propensity score case matching, patients who were transfused did not have higher mortality rates than those who were not transfused. In fact, from the extended Cox proportional hazard analysis, transfused patients had a better survival.
Although a prospective, controlled randomized clinical trial is, of course, the optimal means of demonstrating cause and effect, epidemiologic studies with adequate multivariable analysis can provide valuable information. A similar approach has been taken to show that aspirin administration may reduce complications after coronary artery bypass grafting.20 Our study is limited by its observational design. However, the multivariate regression analysis and the propensity score matching yielded similar results. Nevertheless, other confounders not reported in our study could have contributed to either more beneficial or more deleterious effects of erythrocyte transfusion. Although the SOAP study was primarily designed to study the incidence of sepsis, the current substudy was one of several analyses planned before the SOAP study was conducted.
Importantly, we used exactly the same approach in the ABC study6 but found different results. The study protocols between these two studies (ABC and SOAP) were very similar, with the only possible difference being that Eastern European countries contributed a minority of patients in the current study and none in the ABC study; owing to the small numbers of patients involved, we do not believe this could have influenced the results substantially. Data analysis was also performed in exactly the same way.
Hence, differences are likely to be due to differences in the blood transfusions per se  . Heightened awareness of the possible risks of blood transfusion have led to changes in blood preparation so that blood transfusions are perhaps safer today in terms of viral transmission than they were a decade ago. Leukodepletion, which may reduce some of the negative immunosuppressive effects of transfusions, has also been widely implemented. The leukocyte component of transfused blood has been implicated in some of the adverse effects associated with blood transfusion, including transfusion-related immunomodulation and transfusion-related acute lung injury. Leukoreduction is a process in which the leukocyte component is deliberately reduced by centrifugation or filtration, thus potentially limiting some of these effects. Leukoreduction is also effective in reducing the transmission of cell-associated viruses.21 
In the ABC study, few data were collected regarding leukodepleted blood (46% of centers indicated that they used leukodepleted blood most of the time, 35% used it some of the time, and 19% never used it), showing simply that it was not widely used in Europe at that time. In the current SOAP study, 76% of centers who replied were routinely using leukodepleted blood, demonstrating that leukodepleted blood is now much more commonly used across Europe. It is interesting to speculate that this may account for the differences between the previous ABC study and the current SOAP study. Hebert et al.  22 performed a study comparing patient outcomes before and after introduction of routine blood leukodepletion and noted reduced in-hospital mortality rates after the introduction of leukodepletion compared with the control period (6.19 vs.  7.03%; P  = 0.04). Other studies have also shown improved survival rates with transfusion of leukodepleted compared with standard blood,23–25 although these findings have not been universal.26 
In conclusion, this study suggests that blood transfusions may no longer be associated with increased mortality rates and may be associated with improved survival. A randomized controlled study similar to that conducted by Hebert et al.  between 1994 and 1997 is urgently needed to confirm these findings.
Appendix: Participants by Country#
Austria:  Georg Delle Karth, Department of Cardiology, University Hospital of Vienna; Volker Draxler, Department of Anesthesiology and Intensive Care Medicine, LKH Steyr; Gottfried Filzwieser, Department of Anesthesiology and Intensive Care Medicine, LKH-Deutschlandsberg; Werner Heindl, Department of Intensive Care Medicine, Otto Wagner Spital Hospital, Vienna; Gerhard Kellner, T. Bauer, Department of Anesthesiology and Intensive Care Medicine, Krems Hospital, Donau; Kurt Lenz, Department of Medicine, Barmherzige Bruede, Linz; Edmund Rossmann, Department of Anesthesiology, KH Floridsdorf, Vienna; Christian Wiedermann, Department of Internal Medicine, University Hospital of Innsbruck. Belgium:  Patrick Biston, Department of Intensive Care Medicine, Centre Hospitalier Universitaire (CHU) of Charleroi; Didier Chochrad, Department of Intensive Care Medicine, Hôpitaux Iris Sud, Brussels; Vincent Collin, Department of Intensive Care Medicine, Clinique Europe Site St. Michel, Brussels; Pierre Damas, Department of Intensive Care Medicine, CHU, Liège; Johan Decruyenaere, Eric Hoste, Department of Intensive Care Medicine, University Hospital, Ghent; Jacques Devriendt, Department of Intensive Care Medicine, CHU Brugmann, Brussels; Benoît Espeel, Department of Intensive Care Medicine, Centre Hospitalier Jolimont-Lobbes, Haine St. Paul; Vincent Fraipont, Department of Intensive Care Medicine, CHR Citadelle, Liege; Etienne Installe, Department of Intensive Care Medicine, UCL Mont-Godinne, Yvoir; Manu Malbrain, Department of Intensive Care Medicine, ACZA Campus Stuivenberg, Antwerp; Guy Nollet, Department of Intensive Care Medicine, OLV Ziekenhuis, Aalst; Jean-Charles Preiser, Department of Intensive Care Medicine, RHMS, Tournai; Jan Raemaekers, Department of Intensive Care Medicine, AZ St. Augustinus, Wilrijk; Alain Roman, Department of Intensive Care Medicine, CHU Saint-Pierre, Brussels; Marc Simon, Department of Intensive Care Medicine, Cliniques du Sud-Luxembourg, Arlon; Hebert Spapen, Department of Intensive Care Medicine, Academic Hospital Vrije Universiteit, Brussels; Walter Swinnen, Department of Intensive Care Medicine, AZ Sint-Blasius, Dendermonde; Frédéric Vallot, Department of Intensive Care Medicine, Clinique Notre-Dame, Tournai; Jean-Louis Vincent, Department of Intensive Care Medicine, Erasme University Hospital, Brussels. Czech Republic:  Ivan Chytra, Department of Anesthesiology and Intensive Care Medicine, University Hospital, Plzen; Lukas Dadak, Department of Anesthesiology and Intensive Care Medicine, St. Anne’s University Hospital, Brno; Ivan Herold, Department of Anesthesiology and Intensive Care Medicine, Klaudians, Mlada Boleslav; Ferdinand Polak, Department of Intensive Care Medicine, General Faculty Hospital, Prague; Martin Sterba, Department of Anesthesiology and Intensive Care Medicine, City Hospital, Ostrava. Denmark:  Morten Bestle, Department of Intensive Care Medicine, Gentofte Hospital, University of Copenhagen; Kurt Espersen, Department of Intensive Care Medicine, Rigshospitalet, Copenhagen; Henrik Guldager, Department of Intensive Care Medicine, Amager Hospital, Copenhagen; Karen-Lise Welling, Department of Thoracic Anesthesia, Rigshospitalet, University of Copenhagen. Finland:  Dag Nyman, Department of Intensive Care Medicine, Aland Central Hospital, Mariehamn; Esko Ruokonen, Department of Intensive Care Medicine, University Hospital, Kuopio; Kari Saarinen, Department of Intensive Care Medicine, Seinajoki Central Hospital, Seinajoki. France:  Djillali Annane, Medical Intensive Care Unit, Raymond Poincare, Garches; Philippe Catogni, Department of Intensive Care Medicine, Institut Gustave Roussy, Villejuif; Gabriel Colas, Department of Surgical Intensive Care Medicine, Jacques Monod Hospital, Le Havre; François Coulomb, Department of Intensive Care Medicine, CH Victor Jousselin, Dreux; Rene Dorne, Cardiac Intensive Care Unit, Hôpital St. Joseph and St. Luc, Lyon; Maite Garrouste, Department of Intensive Care Medicine, Hôpital Saint Joseph, Paris; Christian Isetta, Department of Cardiac Surgery, Hôpital Pasteur, Nice; Jérome Larché, Medical Intensive Care Unit, CHU Brabois, Vandoeuvre Les Nancy; Jean-Roger LeGall, Medical Intensive Care Unit, Hôpital Saint Louis, Paris; Henry Lessire, Cardiovascular and Thoracic Intensive Care Medicine, CHU, Grenoble; Yannick Malledant, Department of Surgical Intensive Care, CHU Pontchaillou, Rennes; Philippe Mateu, Department of Intensive Care Medicine, Hôpital des Hauts Clos, Troyes; Michel Ossart, Department of Intensive Care Medicine, CHU, Amiens; Didier Payen, Department of Anesthesiology and Intensive Care Medicine, Hôpital Lariboisière, Paris; Pascal Schlossmacher, Department of Intensive Care Medicine, CHD Félix Gyuon of Saint Denis, La Reunion; Jean-François Timsit, Department of Intensive Care Medicine, Hôpital Bichat, Paris; Stéphane Winnock, Department of Intensive Care Medicine, Hôpital Saint André, Bordeaux; Jean-Pierre Sollet, Department of Intensive Care Medicine, Hôpital Victor Dupouy, Argentueil; Laurent Mallet, Department of Intensive Care Medicine, CH, Auch; Peter Maurer, Medical Intensive Care Unit, CHU Nancy-Brabois, Vandoeuvre; Jean-Michel Sab, Department of Intensive Care Medicine, CH William Morey, Chalon. Germany:  Gueclue Aykut, Department of Anesthesiology, University Hospital, Heidelberg; Frank Brunkhorst, Department of Anesthesiology and Intensive Care Medicine, Friedrich Schiller University, Jena; Rainer Gatz, Department of Anesthesiology, Hans Sushemihl Krankenhaus, Emden; Herwig Gerlach, Department of Anesthesiology and Intensive Care Medicine, Vivantes-Klinikum Neukoelln, Berlin; Dietrich Henzler, Department of Anesthesiology, University Hospital, Aachen; Hans-Bernd Hopf, Department of Anesthesia and Perioperative Medicine, Kreisklinik, Langen-Seligenstadt; Hilmar Hueneburg, Department of Anesthesiology and Intensive Care Medicine, GKH, Bonn; Waheed Karzai, Department of Anesthesiology and Intensive Care Medicine, Zentralklinik, Bad Berka; Ansgar Keller, Department of Anesthesiology, Neuwerk, Moenchengladbach; Uwe Kuhlmann, Department of Nephrology, Philipps University, Marburg; Julia Langgartner, Department of Internal Medicine, University Hospital, Regensburg; Michael Lauterbach, Department of Intensive Care Medicine, University Hospital, Mainz; Cornelia Manhold, Department of Intensive Care Medicine, ZKH Links der Weser, Bremen; Axel Nierhaus, Department of Anesthesiology, University Clinic Hamburg-Eppendorf; Max Ragaller, Department of Anesthesiology and Intensive Care Medicine, University Hospital Carl Gustav Carus, Dresden; Bernd Reith, Department of Surgery, University of Wuerzburg; Tobias Schuerholz, Department of Anesthesiology, Hannover Medical School, Hannover; Claudia Spies, Department of Anesthesiology and Intensive Care Medicine, Universitätsklinikum Charité Campus Mitte, Berlin; Reimund Stögbauer, Department of Intensive Care Medicine, Bethanien Hospital, Moers; Juergen Unterburger, Department of Anesthesiology and Intensive Care Medicine, Schongau Hospital, Schongau. Greece:  Phyllis-Maria Clouva-Molyvdas, Department of Intensive Care Medicine, Thriassio Hospital, Athens; George Giokas, Department of Intensive Care Medicine, Sismanoglion General Hospital, Athens; Eleni Ioannidou, Department of Intensive Care Medicine, Accident Hospital, Athens; Alexandra Lahana, Department of Intensive Care Medicine, G. Papanikolaou General Hospital, Thessaloniki; Alexandros Liolios, Department of Intensive Care Medicine, Agios Demetrios, Thessaloniki; Katerina Marathias, Department of Intensive Care Medicine, Onassis Cardiac Surgery Center, Athens; George Nakos, Department of Intensive Care Medicine, University Hospital, Ioannina; Antonia Tasiou, Department of Intensive Care Medicine, Tzanio Hospital, Athens; Hercules Tsangaris, Department of Intensive Care Medicine, Athens General Hospital Gennimatas, Athens. Hungary:  Peter Tamasi, Department of Intensive Care Medicine, Peterfy Hospital of Budapest. Ireland:  Brian Marsh, Department of Intensive Care Medicine, Mater Hospital, Dublin; Michael Power, Department of Anesthesiology, Beaumont Hospital, Dublin. Israel:  Charles Sprung, Department of Anesthesiology and Intensive Care Medicine, Hadassah Hebrew University Medical Center. Italy:  Bonizella Biagioli, Department of Intensive Care Medicine, Azienda Ospedaliera (AO) Senese o Siena; Franco Bobbio Pallavicini, Department of Intensive Care Medicine, S. Martino, Genoa; Carlo Capra, Department of Anesthesiology and Intensive Care Medicine, Osp Regionale, Saronno; Francesco Della Corte, Department of Intensive Care Medicine, Ospedale Maggiore, University A. Avogadro, Novara; Pier Paolo Donadio, Department of Anesthesiology and Intensive Care Medicine, Osp. Molinette, Turin; Abele Donati, Department of Intensive Care Medicine, A.O. Umberto, Ancona; Antonino Giarratano, Department of Anesthesiology and Intensive Care Medicine, Azienda Ospedaliera Universitaria Policlinico, Palermo; Daniela Giudici, Department of Anesthesiology and Intensive Care Medicine, H San Raffaele, Milan; Stefano Greco, Department of Anesthesiology and Intensive Care Medicine, Ospedale Di Busto, Arsizio; Alberto Guadagnucci, Department of Anesthesiology and Intensive Care Medicine, Civile Di Massa, Massa; Gaetano Iapichino, Department of Anesthesiology and Intensive Care Medicine, San Paolo, Milan; Sergio Livigni, Department of Intensive Care Medicine, S. Giovanni Bosco, Turin; Gabriella Moise, Department of Anesthesiology and Intensive Care Medicine, Osp. San Giovanni, Sesto; Giuseppe Nardi, Department of Intensive Care Medicine, S. Camillo, Rome; Ettore Panascia, Department of Intensive Care Medicine, Vittorio Emanuele, Catania; Antonio Pesenti, Department of Intensive Care Medicine, Azienda Ospedaliera S. Gerardo dei Tintori, Monza; Mario Pizzamiglio, Department of Intensive Care Medicine, Hospital of Piacenza, Piacenza; V. Marco Ranieri, Department of Anesthesiology and Intensive Care Medicine, Universita di Torino-Ospedale S. Giovanni Battista, Turin; Roberto Rosi, Department of Intensive Care Medicine, Policlinico Le Scotte, Siena; Alberto Sicignano, Department of Intensive Care Medicine, Ospedale Maggiore Policlinico IRCCS, Milan; Maurizio Solca, Department of Anesthesiology and Intensive Care Medicine, A. Uboldo, Cernusco Sul Naviglio; Giorgio Tulli, Department of Anesthesiology and Intensive Care Medicine, San Giovanni Di Dio, Florence; Giuliano Vignali, Department of Anesthesiology and Intensive Care Medicine, Hospital Carrara, Massa; Italo Volpe Rinonapoli, Department of Emergency Medicine, San Giovanni, Rome. The Netherlands:  Michel Barnas, Department of Intensive Care Medicine, Boven IJ Ziekenhuis, Amsterdam; Ernst De Bel, Department of Intensive Care Medicine, UMC St. Radboud, Nijmegen; Anne-Cornelie De Pont, Department of Intensive Care Medicine, Academic Medical Center, Amsterdam; Johan Groeneveld, Department of Intensive Care Medicine, VUMC, Amsterdam; Maarten Nijsten, Department of Surgical Intensive Care, Groningen University Hospital, Groningen; Liang-Hai Sie, Department of Intensive Care Medicine, Waterlandziekenhuis, Purmerend; Durk Zandstra, Department of Intensive Care Medicine, OLVG, Amsterdam. Norway:  Svein Harboe, Department of Intensive Care Medicine, Sentralsjukehuset i Rogaland, Stavanger; Svante Lindén, Department of Intensive Care Medicine, Sykehuset Østfold, Fredrikstad; Renata Lovstad, Department of Anesthesiology, Aker University Hospital, Oslo; Harald Moen, Department of Intensive Care Medicine, Ulleval University Hospital, Oslo; Nils Smith-Erichsen, Department of Anesthesiology, Akershus University Hospital, Nordbyhagen. Poland:  Andrzej Piotrowski, Department of Intensive Care Medicine, Pediatric University Hospital, Lodz; Ewa Karpel, Department of Intensive Care Medicine, Medical University of Silesia, Katowice. Portugal:  Eduardo Almeida, Department of Intensive Care Medicine, Garcia de Orta, Almada; Rui Moreno, Department of Intensive Care Medicine, Hospital de St. António dos Capuchos, Lisbon; Antonio Pais-De-Lacerda, Department of Intensive Care Medicine, Hospital de Santa Maria, Lisbon; José Artur Paiva, Department of Intensive Care Medicine, Hospital S. Joao, Porto; Isabel Serra, Department of Intensive Care Medicine, Fernado Fonseca, Masama; Antonio Pimentel, Department of Intensive Care Medicine, São Teotonio, Viseu. Romania:  Daniela Filipescu, Department of Cardiac Anesthesiology and Intensive Care Medicine, Institute of Cardiovascular Diseases, Bucharest. Serbia and Montenegro:  Ksenija Jovanovic, Department of Anesthesiology and Intensive Care Medicine, Military Medical Academy, Belgrade. Slovakia:  Peter Malik, Department of Intensive Care Medicine, National Institute of Cardiovascular Diseases, Bratislava. Slovenia:  Kosec Lucka, Department of Surgical Intensive Care, General Hospital, Novo Mesto; Gorazd Voga, Department of Intensive Care Medicine, General Hospital, Celje. Spain:  Cesar Aldecoa Alvarez-Santullano, Department of Anesthesiology, Hospital Universitario Rio Hortega, Valladolid; Antonio Artigas, Department of Intensive Care Medicine, Sabadell Hospital, Sabadell; Elizabeth Zavala, Department of Anesthesiology and Surgical Intensive Care, Hospital Clinic, Barcelona; Angels Escorsell, Liver Unit, Hospital Clinic, Barcelona; José Nicolas, Department of Intensive Care Medicine, Hospital Clinic, Barcelona; José Javier Izura Cea, Department of Intensive Care Medicine, Virgen del Camino, Pamplona; Luis Marina, Department of Intensive Care Medicine, Virgen de la Salud, Toledo; Juan Montejo, Department of Intensive Care Medicine, 12 de Octubre, Madrid; Eduardo Palencia, Department of Intensive Care Medicine, Gregorio Maranon, Madrid; Francisco Santos, Department of Intensive Care Medicine, General Universitario, Elche; Rafael Sierra-Camerino, Department of Intensive Care Medicine, Puerta del Mar, Cadiz; Fernando Sipmann, Department of Intensive Care Medicine, Fundación Jiménez Díaz, Madrid. Sweden:  Keld Brodersen, Department of Intensive Care Medicine, Central Hospital, Kristianstad; Jan Haggqvist, Department of Anesthesiology and Intensive Care Medicine, Soder Hospital, Stockholm; Dan Hermansson, Department of Anesthesiology and Intensive Care Medicine, Sunderby Hospital, Luleå Hans Hjelmqvist, Department of Anesthesiology and Intensive Care Medicine, Huddinge University Hospital, Stockholm. Switzerland:  Kuno Heer, Department of Intensive Care Medicine, Kantonsspital, Luzern; Giorgio Loderer, Department of Anesthesiology and Intensive Care Medicine, Hirslanden Klinik Beau-Site, Bern; Marco Maggiorini, Department of Internal Medicine, University Hospital, Zurich; Hervé Zender, Department of Medicine, La Chaux-de-Fonds Hôpital. United Kingdom:  Peter Andrews, Department of Intensive Care Medicine, Western General Hospital of Edinburgh; Balraj Appadu, Department of Anesthesiology and Intensive Care Medicine, Peterborough Hospitals National Health Service Trust, Peterborough; Casiano Barrera Groba, Department of Intensive Care Medicine, University Hospital, Lewisham; Jeremy Bewley, Department of Anesthesiology, Bristol Royal Infirmary, Bristol; Ken Burchett, Department of Intensive Care Medicine, Queen Elizabeth Hospital, Kings Lynn; Philip Chambers, Department of Anesthesiology and Intensive Care Medicine, General Hospital, Milton Keynes; John Coakley, Department of Intensive Care Medicine, Homerton University Hospital, London; Doris Doberenz, Department of Intensive Care Medicine, Charing Cross Hospital, London; Nigel Eastwood, Department of Anesthesiology and Intensive Care Medicine, North Staffordshire Hospital, Stoke on Trent; Andrew Ferguson, Department of Intensive Care Medicine, Antrim Area Hospital, Antrim; Jonathan Fielden, Department of Intensive Care Medicine, Royal Berkshire Hospital, Reading; Jacqueline Gedney, Department of Intensive Care Medicine, The James Cook University Hospital, Middlesbrough; Kevin Gunning, Department of Intensive Care Medicine, Addenbrookes, Cambridge; Dave Harling, Department of Intensive Care Medicine, Rotherham DGH; Stas Jankowski, Department of Anesthesiology and Intensive Care Medicine, St. Helier, Carshalton; David Jayson, Department of Anesthesiology, Southport and Formby Hospital, Southport; Andrew Kilner, Department of Perioperative and Intensive Care Medicine, Freeman, Newcastle upon Tyne; Venketachalam Krishna-Kumar, Department of Anesthesiology, University Hospital of North Tees, Stockton on Tees; Katie Lei, Department of Intensive Care Medicine, St. Thomas Hospital, London; Simon Mackenzie, Department of Intensive Care Medicine, Royal Infirmary, Edinburgh; Peter Macnaughton, Department of Intensive Care Medicine, Derriford Hospital, Plymouth; Gernot Marx, Department of Intensive Care Medicine, Royal Liverpool University Hospital, Liverpool; C. McCulloch, Department of Anesthesiology and Intensive Care, Stirling Royal Infirmary; Paul Morgan, Department of Intensive Care Medicine, University Hospital of Wales, Cardiff; Andy Rhodes, Department of Intensive Care Medicine, St. George’s Hospital, London; Chris Roberts, Department of Intensive Care Medicine, Gloucestershire Royal Hospital, Gloucester; Mark Russell, Department of Intensive Care Medicine, St. Peters, Chertsey; Darell Tupper-Carey, Department of Anesthesiology, James Paget Hospital, Great Yarmouth; Maggie Wright, Department of Intensive Care Medicine, James Paget Hospital, Great Yarmouth; Linda Twohey, Department of Anesthesiology, Kettering General Hospital, Kettering; James Watts, Department of Anesthesiology Burnley DGH, Burnley; Rae Webster, Department of Anesthesiology and Intensive Care Medicine, Northampton General Hospital, Northampton; Dewi Williams, Department of Anesthesiology, Royal Infirmary, Dumfries.
References
Dellinger EP, Anaya DA: Infectious and immunologic consequences of blood transfusion. Crit Care 2004; 8 (suppl 2):S18–23Dellinger, EP Anaya, DA
Blajchman MA: Transfusion immunomodulation or TRIM: What does it mean clinically? Hematology 2005; 10 (suppl 1):208–14Blajchman, MA
Carson JL, Duff A, Poses RM, Berlin JA, Spence RK, Trout R, Noveck H, Strom BL: Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet 1996; 348:1055–60Carson, JL Duff, A Poses, RM Berlin, JA Spence, RK Trout, R Noveck, H Strom, BL
Carson JL, Noveck H, Berlin JA, Gould SA: Mortality and morbidity in patients with very low postoperative Hb levels who decline blood transfusion. Transfusion 2002; 42:812–8Carson, JL Noveck, H Berlin, JA Gould, SA
Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M, Schweitzer I, Yetisir E: A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 1999; 340:409–17Hebert, PC Wells, G Blajchman, MA Marshall, J Martin, C Pagliarello, G Tweeddale, M Schweitzer, I Yetisir, E
Vincent JL, Baron JF, Reinhart K, Gattinoni L, Thijs L, Webb A, Meier-Hellmann A, Nollet G, Peres-Bota D: Anemia and blood transfusion in critically ill patients. JAMA 2002; 288:1499–507Vincent, JL Baron, JF Reinhart, K Gattinoni, L Thijs, L Webb, A Meier-Hellmann, A Nollet, G Peres-Bota, D
Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot MM, Duh MS, Shapiro MJ: The CRIT Study: Anemia and blood transfusion in the critically ill—current clinical practice in the United States. Crit Care Med 2004; 32:39–52Corwin, HL Gettinger, A Pearl, RG Fink, MP Levy, MM Abraham, E MacIntyre, NR Shabot, MM Duh, MS Shapiro, MJ
Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM: Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma 2003; 54:898–905Malone, DL Dunne, J Tracy, JK Putnam, AT Scalea, TM Napolitano, LM
Palmieri TL, Caruso DM, Foster KN, Cairns BA, Peck MD, Gamelli RL, Mozingo DW, Kagan RJ, Wahl W, Kemalyan NA, Fish JS, Gomez M, Sheridan RL, Faucher LD, Latenser BA, Gibran NS, Klein RL, Solem LD, Saffle JR, Morris SE, Jeng JC, Voigt D, Howard PA, Molitor F, Greenhalgh DG: Effect of blood transfusion on outcome after major burn injury: A multicenter study. Crit Care Med 2006; 34:1602–7Palmieri, TL Caruso, DM Foster, KN Cairns, BA Peck, MD Gamelli, RL Mozingo, DW Kagan, RJ Wahl, W Kemalyan, NA Fish, JS Gomez, M Sheridan, RL Faucher, LD Latenser, BA Gibran, NS Klein, RL Solem, LD Saffle, JR Morris, SE Jeng, JC Voigt, D Howard, PA Molitor, F Greenhalgh, DG
Koch CG, Li L, Duncan AI, Mihaljevic T, Cosgrove DM, Loop FD, Starr NJ, Blackstone EH: Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med 2006; 34:1608–16Koch, CG Li, L Duncan, AI Mihaljevic, T Cosgrove, DM Loop, FD Starr, NJ Blackstone, EH
Rao SV, Jollis JG, Harrington RA, Granger CB, Newby LK, Armstrong PW, Moliterno DJ, Lindblad L, Pieper K, Topol EJ, Stamler JS, Califf RM: Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004; 292:1555–62Rao, SV Jollis, JG Harrington, RA Granger, CB Newby, LK Armstrong, PW Moliterno, DJ Lindblad, L Pieper, K Topol, EJ Stamler, JS Califf, RM
Lacroix J, Hebert PC, Hutchison JS, Hume HA, Tucci M, Ducruet T, Gauvin F, Collet JP, Toledano BJ, Robillard P, Joffe A, Biarent D, Meert K, Peters MJ: Transfusion strategies for patients in pediatric intensive care units. N Engl J Med 2007; 356:1609–19Lacroix, J Hebert, PC Hutchison, JS Hume, HA Tucci, M Ducruet, T Gauvin, F Collet, JP Toledano, BJ Robillard, P Joffe, A Biarent, D Meert, K Peters, MJ
Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, Moreno R, Carlet J, Le Gall JR, Payen D: Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 2006; 34:344–53Vincent, JL Sakr, Y Sprung, CL Ranieri, VM Reinhart, K Gerlach, H Moreno, R Carlet, J Le Gall, JR Payen, D
Le Gall J-R, Lemeshow S, Saulnier F: A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957–63Le Gall, J-R Lemeshow, S Saulnier, F
Vincent JL, Moreno R, Takala J, Willatts S, de Mendonça A, Bruining H, Reinhart CK, Suter PM, Thijs LG: The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Med 1996; 22:707–10Vincent, JL Moreno, R Takala, J Willatts, S de Mendonça, A Bruining, H Reinhart, CK Suter, PM Thijs, LG
ACCP-SCCM Consensus Conference: Definitions of sepsis and multiple organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864–74
Rosenbaum PR, Rubin DB: The central role of the propensity score on observational studies for causal effects. Biometrika 1983; 70:41–55Rosenbaum, PR Rubin, DB
Gum PA, Thamilarasan M, Watanabe J, Blackstone EH, Lauer MS: Aspirin use and all-cause mortality among patients being evaluated for known or suspected coronary artery disease: A propensity analysis. JAMA 2001; 286:1187–94Gum, PA Thamilarasan, M Watanabe, J Blackstone, EH Lauer, MS
Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. Wiley Series in Probability and Mathematical Statistics. Hoboken, New Jersey, John Wiley & Sons, 1980Kalbfleisch, JD Prentice, RL Hoboken, New Jersey John Wiley & Sons
Mangano DT: Aspirin and mortality from coronary bypass surgery. N Engl J Med 2002; 347:1309–17Mangano, DT
Shapiro MJ: To filter blood or universal leukoreduction: What is the answer? Crit Care 2004; 8 (suppl 2):S27–30Shapiro, MJ
Hebert PC, Fergusson D, Blajchman MA, Wells GA, Kmetic A, Coyle D, Heddle N, Germain M, Goldman M, Toye B, Schweitzer I, van Walraven C, Devine D, Sher GD: Clinical outcomes following institution of the Canadian universal leukoreduction program for red blood cell transfusions. JAMA 2003; 289:1941–9Hebert, PC Fergusson, D Blajchman, MA Wells, GA Kmetic, A Coyle, D Heddle, N Germain, M Goldman, M Toye, B Schweitzer, I van Walraven, C Devine, D Sher, GD
Fergusson D, Hebert PC, Lee SK, Walker CR, Barrington KJ, Joseph L, Blajchman MA, Shapiro S: Clinical outcomes following institution of universal leukoreduction of blood transfusions for premature infants. JAMA 2003; 289:1950–6Fergusson, D Hebert, PC Lee, SK Walker, CR Barrington, KJ Joseph, L Blajchman, MA Shapiro, S
Tartter PI, Mohandas K, Azar P, Endres J, Kaplan J, Spivack M: Randomized trial comparing packed red cell blood transfusion with and without leukocyte depletion for gastrointestinal surgery. Am J Surg 1998; 176:462–6Tartter, PI Mohandas, K Azar, P Endres, J Kaplan, J Spivack, M
Bilgin YM, van de Watering LM, Eijsman L, Versteegh MI, Brand R, van Oers MH, Brand A: Double-blind, randomized controlled trial on the effect of leukocyte-depleted erythrocyte transfusions in cardiac valve surgery. Circulation 2004; 109:2755–60Bilgin, YM van de Watering, LM Eijsman, L Versteegh, MI Brand, R van Oers, MH Brand, A
Phelan HA, Sperry JL, Friese RS: Leukoreduction before red blood cell transfusion has no impact on mortality in trauma patients. J Surg Res 2007; 138:32–6Phelan, HA Sperry, JL Friese, RS
Fig. 1. Cumulative hazard of death in the propensity-matched groups (  solid line  = transfusion;  dashed line  = no transfusion). Hazard ratio = 0.73; 95% confidence interval = 0.59–0.90;  P  = 0.004. 
Image Not Available
Fig. 1. Cumulative hazard of death in the propensity-matched groups (  solid line  = transfusion;  dashed line  = no transfusion). Hazard ratio = 0.73; 95% confidence interval = 0.59–0.90;  P  = 0.004. 
×
Table 1. The Logistic Regression Model Used to Calculate the Propensity Score 
Image Not Available
Table 1. The Logistic Regression Model Used to Calculate the Propensity Score 
×
Table 2. Characteristics of the Study Group 
Image Not Available
Table 2. Characteristics of the Study Group 
×
Table 3. Simplified Acute Physiology Score II, ICU and Hospital Mortality According to the Number of Transfused Units during ICU Stay 
Image Not Available
Table 3. Simplified Acute Physiology Score II, ICU and Hospital Mortality According to the Number of Transfused Units during ICU Stay 
×
Table 4. Transfusion Rate and Mean Number of Units Transfused According to ICU Duration of Stay 
Image Not Available
Table 4. Transfusion Rate and Mean Number of Units Transfused According to ICU Duration of Stay 
×
Table 5. Multivariate Cox Proportional Hazard Analysis with Time to In-hospital Mortality at 30 Days as the Dependent Variable 
Image Not Available
Table 5. Multivariate Cox Proportional Hazard Analysis with Time to In-hospital Mortality at 30 Days as the Dependent Variable 
×
Table 6. Patient Characteristics by Transfusion Status for the Propensity-matched Patients 
Image Not Available
Table 6. Patient Characteristics by Transfusion Status for the Propensity-matched Patients 
×