Close to one million red blood cell (RBC) units are transfused annually in Canadian hospitals, with surgical inpatients accounting for up to 44% of transfusions. There is evidence of significant variation in transfusion practice in the operating room (i.e., intraoperative). Although variation is expected based on disease severity and patient preference, inappropriate clinical care due to either under- or over-transfusion likely also contributes to significant variation. Indeed, estimates of unwarranted intraoperative RBC transfusions in the literature range from 19% to 49%, owing partly to a lack of evidence-based consensus on RBC transfusion practice in the OR. Our two systematic reviews have highlighted this gap, demonstrating a lack of evidence from trials or actionable clinical practice guidelines to inform decisions in the OR. Perhaps more importantly, avoidance of blood product exposure is an important patient-prioritized outcome that has yet to be studied empirically in the OR. As such, the observed variation in transfusion practice suggests that transfusion decision-making during surgery represents a clear and important knowledge and evidence gap. Transfusion decision-making in the OR is a complex and dynamic process that we cannot begin to improve without first understanding it. It is influenced by 1) physiologic parameters such as acute blood loss, the effects of general anesthesia, and surgical manipulation. Decision-making is also likely heavily influenced by 2) behavioural factors in the OR (heuristics, team dynamics, institutional culture), for which very little empirical work has been conducted. Finally, the importance of 3) patient input in influencing transfusion decisions is inadequately studied, given the documented disconnect between patient priorities and outcomes used in the medical literature and by clinicians. In this context, the aim of my thesis was to develop an empirical understanding of transfusion decision-making in the OR based on stakeholder perceptions and priorities, informed by an integrated patient engagement process. With this work, I address an important knowledge gap in intraoperative blood transfusion, thereby contributing to efforts to reduce variation in blood transfusion practice in surgery. It is my hope that this work will be influential in informing actionable perioperative tools to optimize blood management including providing both evidence and knowledge gaps for future research.

Introdução: O objetivo deste estudo foi avaliar o efeito de uma estratégia restritiva de transfusão de hemácias comparada a uma estratégia liberal na evolução clínica de pacientes submetidos à cirurgia cardíaca com circulação extracorpórea. Métodos: Estudo prospectivo, randomizado, e controlado. Foram randomizados 512 pacientes para uma estratégia liberal de transfusão (hematócrito 30%) ou restritiva (hematócrito 24%). Resultados: Os pacientes apresentaram idade média de 60,7 ± 12,5 anos no grupo liberal e 58,6 ± 12,5 anos no grupo restritivo. Em ambos, houve predomínio de pacientes com fração de ejeção normal, mas 13% dos pacientes do grupo liberal e 15% do grupo restritivo apresentavam fração de ejeção abaixo de 40%. A mortalidade ou a morbidade grave em 30 dias foi semelhante nos dois grupos (10% na estratégia liberal e 11% na estratégia restritiva, P=0,518). Não houve diferença entre os grupos em relação às taxas de complicações secundárias. A média da concentração de hemoglobina foi 10,5 0,9 g/dL no grupo liberal e 9,1 1,2 gdL no grupo restritivo (P<0,001). No grupo liberal, 198 pacientes (78%) receberam transfusão de hemácias, e no grupo restritivo 118 pacientes (47%). Independente da estratégia utilizada, o número de transfusão de hemácias foi fator preditor independente para a ocorrência de complicações clínicas graves ou morte em 30 dias (OR=1,21; IC 95%=1,1-1,4, P= 0,002). Conclusão: A estratégia restritiva de transfusão foi tão segura quanto à estratégia liberal em pacientes submetidos à cirurgia cardíaca com circulação extracorpórea. Cada unidade de concentrado de hemácias administrada foi associada a aumento de 1,2 vezes no risco de mortalidade em 30 dias / Introduction: The aim of this study was to evaluate whether a restrictive strategy of red blood cells transfusion is as safe as a liberal one with respect to clinical outcomes. Methods: In this prospective, randomized, controlled clinical trial we randomly assigned 512 patients after cardiac surgery to a liberal strategy of transfusion (to maintain a hematocrit at least at 30%) or to a restrictive one (to maintain a hematocrit at least at 24%). Results: The mean age of patients was 60.7 ± 12.5 year-old in liberal group vs. 58.6 ± 12.5 year-old in restrictive group. In both groups, most patients had normal left ventricular function, but 13% of patients in the liberal group and 15% in the restrictive one presented ejection fraction of less than 40%. The primary end-point - a composite endpoint of 30 day all cause mortality or severe morbidity was similar between groups (10% in the liberal-strategy group vs. 11% in the restrictive-strategy group) (P=0.518). Also, there were no differences between groups with respect to the rates of other complications. Hemoglobin concentrations were maintained at a level of 10.5 ± 0.9 g/dL in the liberal group and 9.1 ± 1.2 g/dL in the restrictive group (P<0.001). A total of 198 patients (78%) in the liberal group and 118 patients (47%) in the restrictive group received a blood transfusion (P<0.001). Independently of the strategy group, the number of transfused RBC units was an independent risk factor for the occurrence of several clinical complications or death at 30 days (HR = 1.21; CI 95%=1.1-1.4, P=0.002). Conclusions: A restrictive transfusion strategy was as safe as a liberal strategy in patients undergoing cardiac surgery. For each RBC unit transfused, transfusion was independently associated with a 1.2-fold higher risk of death at 30 days

OBJETIVOS: O objetivo do estudo foi avaliar se a transfusão de hemácias no intraoperatório de cirurgia cardíaca com circulação extracorpórea está associada a complicações clínicas incluindo choque cardiogênico, arritmia, insuficiência renal aguda, isquemia miocárdica, choque séptico, necessidade de reintubação orotraqueal, acidente vascular cerebral ou mortalidade durante a internação hospitalar. DESENHO: Estudo clínico de coorte retrospectivo e unicêntrico com escore de propensão, realizado no Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. PACIENTES: Pacientes adultos submetidos à cirurgia cardíaca eletiva com circulação extracorpórea no período de janeiro de 2004 a dezembro de 2008. DESFECHO PRIMÁRIO: Complicações clínicas durante a internação hospitalar (choque cardiogênico, arritmia, insuficiência renal aguda, isquemia miocárdica, choque séptico, necessidade de reintubação orotraqueal, acidente vascular cerebral ou mortalidade hospitalar). DESFECHO SECUNDÁRIO: 1- Avaliar o efeito da transfusão de hemácias no intraoperatório no tempo livre de inotrópicos e vasopressores, tempo de ventilação mecânica e tempo de permanência na UTI e internação hospitalar. 2- Avaliar o efeito do número das unidades de hemácias transfundidas no intraoperatório na ocorrência de mortalidade hospitalar, choque cardiogênico, arritmia, isquemia miocárdica, choque séptico, acidente vascular cerebral e reintubação orotraqueal. 3- Avaliar o efeito da anemia à admissão e durante internação hospitalar na ocorrência de complicações pós-operatórias. INTERVENÇÃO: Não houve intervenção. RESULTADOS: Foram incluídos 2851 pacientes na análise final, dos quais 1471(51,6%) foram expostos a transfusão de hemácias e 1380 (48,4%) não receberam transfusão no intraoperatório. Os pacientes transfundidos apresentaram maior incidência das seguintes complicações: mortalidade (2,1% vs 0,4%, P < 0,001), insuficiência renal aguda (9,1% vs 3,9%, P<0,001), reintubação orotraqueal (3,8% vs 1,4%, P < 0,001) e choque séptico (2,2% vs 0,4%, P < 0,001). Os pacientes transfundidos também apresentaram maior tempo de internação hospitalar [16 dias (12-23) vs 13 dias (9-18), P < 0,001] e em unidade de terapia intensiva [3 dias (2-6) vs 2 dias (2-4), P < 0,001]. A concentração da hemoglobina menor que 9 g/dL ocorreu em 1847 pacientes (64,7%) durante a internação hospitalar e foi associada a maior risco de insuficiência renal aguda e de acidente vascular cerebral. O escore de propensão identificou 588 pacientes pareados em relação à exposição à transfusão, e essa análise demonstrou que a transfusão intraoperatória de hemácias não aumentou a ocorrência de complicações no período de internação hospitalar. Contudo a transfusão de 4 ou mais unidades de hemácias está associada a maior ocorrência de mortalidade hospitalar, choque cardiogênico e IRA, maior incidência de reintubação orotraqueal, choque séptico e AVC. Além de uma relação direta entre as unidades de hemácias transfundidas e a ocorrência de morte. CONCLUSÃO: Esse estudo observacional demonstrou que a anemia é frequentemente detectada no pós-operatório de cirurgia cardíaca, e está associada a maior incidência de complicações. Além disso, a transfusão de hemácias no intraoperatório não modifica a ocorrência das complicações pós-operatórias em pacientes submetidos a cirurgia cardíaca. No entanto a transfusão de 4 ou mais hemácias está associada a maior incidência de complicações clínicas, além de uma relação dose-dependente. Estratégias como detecção precoce de anemia e emprego de técnicas alternativas à transfusão no manejo devem ser estimuladas no ambiente perioperatório / OBJECTIVE: The objective of this study was to evaluate whether the transfusion of red blood cells in the intraoperative cardiac surgery with extracorporeal circulation is associated with complications after cardiac surgery. DESIGN: A retrospective cohort study with a propensity score analysis, performed at Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. PATIENTS: Adult patients undergoing elective cardiac surgery with cardiopulmonary bypass in the period of January to 2008 December. PRIMARY OUTCOME: Clinical complications during hospital stay (cardiogenic shock, arrhythmia, cardiogenic shock, acute kidney injury, myocardial ischemia, septic shock, tracheal reintubation, stroke or hospital mortality). SECONDARY OUTCOME: 1- Evaluate the effect of intraoperative red blood cell transfusion in inotropic and vasopressor free time, mechanical ventilation time, length of ICU stay and hospital stay. 2- Evaluate the effect of the number of units of transfused red blood cells intraoperatively on the occurrence of hospital mortality, cardiogenic shock, arrhythmia, myocardial ischemia, septic shock, stroke and orotracheal reintubation. 3- Evaluate the effect of anemia on admission and during hospitalization in the occurrence of postoperative complications. RESULTS: In the final analysis, 2851 patients were included. Of these patients, 1471(51.6%) were exposed to red blood cell transfusion (RBC) and 1380 (48.4%) were not exposed to RBC during intraoperative. Transfused patients had higher incidence of the following complications: mortality (2.1% vs. 0.4%, P < 0.001), acute kidney injury (9.1% vs. 3.9%, P < 0,001), tracheal reintubation (3.8% vs. 1.4%, P < 0.001) and septic shock (2.2% vs. 0.4%, P < 0.001). Transfused patients also had a longer length of hospital stay [16 days (12-23) vs. 13 days (9-18), P<0.001] and prolonged intensive care unit stay [3 days (2-6) vs. 2 days (2-4), P < 0.001]. Hemoglobin lower than 9 g/dL was found in 1847 patients (64.7%) during hospital stay and was associated to a higher risk of acute kidney injury and stroke. The propensity score identified 588 paired patients in relation to transfusion exposure, and this analysis demonstrated that intraoperative transfusion of red blood cells did not increase the occurrence of complications during hospitalization. However, transfusion of 4 or more units of red blood cells is associated with a higher occurrence of hospital mortality, cardiogenic shock and acute renal failure, a higher incidence of orotracheal reintubation, septic shock and stroke. In addition to a direct relationship between the units of transfused red blood cells and the occurrence of death. CONCLUSIONS: This observational study demonstrated that anemia is frequently detected in the postoperative period of cardiac surgery, and is associated with a higher incidence of complications. In addition, red blood cell transfusion in the intraoperative does not modify the occurrence of postoperative complications in patients undergoing cardiac surgery. However, transfusion of 4 or more erythrocytes is associated with a higher incidence of clinical complications, in addition to a dose-dependent relationship. Strategies such as early detection of anemia and use of alternative techniques to transfusion in management should be stimulated in the perioperative environment

Introdução: O objetivo deste estudo foi avaliar o efeito de uma estratégia restritiva de transfusão de hemácias comparada a uma estratégia liberal na evolução clínica de pacientes submetidos à cirurgia cardíaca com circulação extracorpórea. Métodos: Estudo prospectivo, randomizado, e controlado. Foram randomizados 512 pacientes para uma estratégia liberal de transfusão (hematócrito 30%) ou restritiva (hematócrito 24%). Resultados: Os pacientes apresentaram idade média de 60,7 ± 12,5 anos no grupo liberal e 58,6 ± 12,5 anos no grupo restritivo. Em ambos, houve predomínio de pacientes com fração de ejeção normal, mas 13% dos pacientes do grupo liberal e 15% do grupo restritivo apresentavam fração de ejeção abaixo de 40%. A mortalidade ou a morbidade grave em 30 dias foi semelhante nos dois grupos (10% na estratégia liberal e 11% na estratégia restritiva, P=0,518). Não houve diferença entre os grupos em relação às taxas de complicações secundárias. A média da concentração de hemoglobina foi 10,5 0,9 g/dL no grupo liberal e 9,1 1,2 gdL no grupo restritivo (P<0,001). No grupo liberal, 198 pacientes (78%) receberam transfusão de hemácias, e no grupo restritivo 118 pacientes (47%). Independente da estratégia utilizada, o número de transfusão de hemácias foi fator preditor independente para a ocorrência de complicações clínicas graves ou morte em 30 dias (OR=1,21; IC 95%=1,1-1,4, P= 0,002). Conclusão: A estratégia restritiva de transfusão foi tão segura quanto à estratégia liberal em pacientes submetidos à cirurgia cardíaca com circulação extracorpórea. Cada unidade de concentrado de hemácias administrada foi associada a aumento de 1,2 vezes no risco de mortalidade em 30 dias / Introduction: The aim of this study was to evaluate whether a restrictive strategy of red blood cells transfusion is as safe as a liberal one with respect to clinical outcomes. Methods: In this prospective, randomized, controlled clinical trial we randomly assigned 512 patients after cardiac surgery to a liberal strategy of transfusion (to maintain a hematocrit at least at 30%) or to a restrictive one (to maintain a hematocrit at least at 24%). Results: The mean age of patients was 60.7 ± 12.5 year-old in liberal group vs. 58.6 ± 12.5 year-old in restrictive group. In both groups, most patients had normal left ventricular function, but 13% of patients in the liberal group and 15% in the restrictive one presented ejection fraction of less than 40%. The primary end-point - a composite endpoint of 30 day all cause mortality or severe morbidity was similar between groups (10% in the liberal-strategy group vs. 11% in the restrictive-strategy group) (P=0.518). Also, there were no differences between groups with respect to the rates of other complications. Hemoglobin concentrations were maintained at a level of 10.5 ± 0.9 g/dL in the liberal group and 9.1 ± 1.2 g/dL in the restrictive group (P<0.001). A total of 198 patients (78%) in the liberal group and 118 patients (47%) in the restrictive group received a blood transfusion (P<0.001). Independently of the strategy group, the number of transfused RBC units was an independent risk factor for the occurrence of several clinical complications or death at 30 days (HR = 1.21; CI 95%=1.1-1.4, P=0.002). Conclusions: A restrictive transfusion strategy was as safe as a liberal strategy in patients undergoing cardiac surgery. For each RBC unit transfused, transfusion was independently associated with a 1.2-fold higher risk of death at 30 days

OBJETIVOS: O objetivo do estudo foi avaliar se a transfusão de hemácias no intraoperatório de cirurgia cardíaca com circulação extracorpórea está associada a complicações clínicas incluindo choque cardiogênico, arritmia, insuficiência renal aguda, isquemia miocárdica, choque séptico, necessidade de reintubação orotraqueal, acidente vascular cerebral ou mortalidade durante a internação hospitalar. DESENHO: Estudo clínico de coorte retrospectivo e unicêntrico com escore de propensão, realizado no Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. PACIENTES: Pacientes adultos submetidos à cirurgia cardíaca eletiva com circulação extracorpórea no período de janeiro de 2004 a dezembro de 2008. DESFECHO PRIMÁRIO: Complicações clínicas durante a internação hospitalar (choque cardiogênico, arritmia, insuficiência renal aguda, isquemia miocárdica, choque séptico, necessidade de reintubação orotraqueal, acidente vascular cerebral ou mortalidade hospitalar). DESFECHO SECUNDÁRIO: 1- Avaliar o efeito da transfusão de hemácias no intraoperatório no tempo livre de inotrópicos e vasopressores, tempo de ventilação mecânica e tempo de permanência na UTI e internação hospitalar. 2- Avaliar o efeito do número das unidades de hemácias transfundidas no intraoperatório na ocorrência de mortalidade hospitalar, choque cardiogênico, arritmia, isquemia miocárdica, choque séptico, acidente vascular cerebral e reintubação orotraqueal. 3- Avaliar o efeito da anemia à admissão e durante internação hospitalar na ocorrência de complicações pós-operatórias. INTERVENÇÃO: Não houve intervenção. RESULTADOS: Foram incluídos 2851 pacientes na análise final, dos quais 1471(51,6%) foram expostos a transfusão de hemácias e 1380 (48,4%) não receberam transfusão no intraoperatório. Os pacientes transfundidos apresentaram maior incidência das seguintes complicações: mortalidade (2,1% vs 0,4%, P < 0,001), insuficiência renal aguda (9,1% vs 3,9%, P<0,001), reintubação orotraqueal (3,8% vs 1,4%, P < 0,001) e choque séptico (2,2% vs 0,4%, P < 0,001). Os pacientes transfundidos também apresentaram maior tempo de internação hospitalar [16 dias (12-23) vs 13 dias (9-18), P < 0,001] e em unidade de terapia intensiva [3 dias (2-6) vs 2 dias (2-4), P < 0,001]. A concentração da hemoglobina menor que 9 g/dL ocorreu em 1847 pacientes (64,7%) durante a internação hospitalar e foi associada a maior risco de insuficiência renal aguda e de acidente vascular cerebral. O escore de propensão identificou 588 pacientes pareados em relação à exposição à transfusão, e essa análise demonstrou que a transfusão intraoperatória de hemácias não aumentou a ocorrência de complicações no período de internação hospitalar. Contudo a transfusão de 4 ou mais unidades de hemácias está associada a maior ocorrência de mortalidade hospitalar, choque cardiogênico e IRA, maior incidência de reintubação orotraqueal, choque séptico e AVC. Além de uma relação direta entre as unidades de hemácias transfundidas e a ocorrência de morte. CONCLUSÃO: Esse estudo observacional demonstrou que a anemia é frequentemente detectada no pós-operatório de cirurgia cardíaca, e está associada a maior incidência de complicações. Além disso, a transfusão de hemácias no intraoperatório não modifica a ocorrência das complicações pós-operatórias em pacientes submetidos a cirurgia cardíaca. No entanto a transfusão de 4 ou mais hemácias está associada a maior incidência de complicações clínicas, além de uma relação dose-dependente. Estratégias como detecção precoce de anemia e emprego de técnicas alternativas à transfusão no manejo devem ser estimuladas no ambiente perioperatório / OBJECTIVE: The objective of this study was to evaluate whether the transfusion of red blood cells in the intraoperative cardiac surgery with extracorporeal circulation is associated with complications after cardiac surgery. DESIGN: A retrospective cohort study with a propensity score analysis, performed at Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. PATIENTS: Adult patients undergoing elective cardiac surgery with cardiopulmonary bypass in the period of January to 2008 December. PRIMARY OUTCOME: Clinical complications during hospital stay (cardiogenic shock, arrhythmia, cardiogenic shock, acute kidney injury, myocardial ischemia, septic shock, tracheal reintubation, stroke or hospital mortality). SECONDARY OUTCOME: 1- Evaluate the effect of intraoperative red blood cell transfusion in inotropic and vasopressor free time, mechanical ventilation time, length of ICU stay and hospital stay. 2- Evaluate the effect of the number of units of transfused red blood cells intraoperatively on the occurrence of hospital mortality, cardiogenic shock, arrhythmia, myocardial ischemia, septic shock, stroke and orotracheal reintubation. 3- Evaluate the effect of anemia on admission and during hospitalization in the occurrence of postoperative complications. RESULTS: In the final analysis, 2851 patients were included. Of these patients, 1471(51.6%) were exposed to red blood cell transfusion (RBC) and 1380 (48.4%) were not exposed to RBC during intraoperative. Transfused patients had higher incidence of the following complications: mortality (2.1% vs. 0.4%, P < 0.001), acute kidney injury (9.1% vs. 3.9%, P < 0,001), tracheal reintubation (3.8% vs. 1.4%, P < 0.001) and septic shock (2.2% vs. 0.4%, P < 0.001). Transfused patients also had a longer length of hospital stay [16 days (12-23) vs. 13 days (9-18), P<0.001] and prolonged intensive care unit stay [3 days (2-6) vs. 2 days (2-4), P < 0.001]. Hemoglobin lower than 9 g/dL was found in 1847 patients (64.7%) during hospital stay and was associated to a higher risk of acute kidney injury and stroke. The propensity score identified 588 paired patients in relation to transfusion exposure, and this analysis demonstrated that intraoperative transfusion of red blood cells did not increase the occurrence of complications during hospitalization. However, transfusion of 4 or more units of red blood cells is associated with a higher occurrence of hospital mortality, cardiogenic shock and acute renal failure, a higher incidence of orotracheal reintubation, septic shock and stroke. In addition to a direct relationship between the units of transfused red blood cells and the occurrence of death. CONCLUSIONS: This observational study demonstrated that anemia is frequently detected in the postoperative period of cardiac surgery, and is associated with a higher incidence of complications. In addition, red blood cell transfusion in the intraoperative does not modify the occurrence of postoperative complications in patients undergoing cardiac surgery. However, transfusion of 4 or more erythrocytes is associated with a higher incidence of clinical complications, in addition to a dose-dependent relationship. Strategies such as early detection of anemia and use of alternative techniques to transfusion in management should be stimulated in the perioperative environment

Les transfusions sanguines sont fréquemment employées pour corriger l’anémie secondaire à une arthroplastie de la hanche ou du genou. Il n’y a cependant pas consensus sur les indications de transfuser. La tendance actuelle est d’utiliser une stratégie transfusionnelle restrictive (soit un seuil de 75-80 g/L d’hémoglobine) mais les conséquences d’une telle pratique sur la récupération fonctionnelle et la qualité de vie des patients sont mal connues. Dans un premier temps, nous avons caractérisé la pratique transfusionnelle au Centre hospitalier de l’Université de Montréal (CHUM). Notre hypothèse était que, devant l’imprécision des recommandations, la pratique transfusionnelle serait variable. Une étude rétrospective de 701 dossiers de patients ayant subi une arthroplastie de la hanche ou du genou a été réalisée. Nous avons observé que les transfusions étaient utilisées de la même façon dans les trois hôpitaux et que les médecins basaient leur décision de transfuser principalement sur un seul chiffre, la concentration d’hémoglobine, adoptant une stratégie restrictive. Soixante-six pourcent des patients avaient une concentration d’hémoglobine inférieure à 100 g/L au départ de l’hôpital. Dans un deuxième temps, nous avons évalué l’impact de cette anémie postopératoire sur la récupération fonctionnelle et la qualité de vie des patients. Notre hypothèse était qu’il existe une concentration d’hémoglobine en dessous de laquelle celles-ci sont atteintes. Une étude de cohorte prospective et observationnelle a été menée chez 305 patients regroupés selon leur concentration d’hémoglobine postopératoire. Les groupes d’hémoglobine (≤ 80, 81-90, 91-100 et > 100 g/L) étaient similaires dans l’évolution de la distance de marche en six minutes, de l’évaluation de l’effort fourni, de la force de préhension et des scores de qualité de vie. L’anémie modérée n’est donc pas associée à une atteinte de la récupération fonctionnelle et de la qualité de vie à court terme. D’autres études devront déterminer les conséquences à long terme d’une stratégie transfusionnelle restrictive sur ces patients. / Red blood cell transfusions are frequently used to treat anemia after total hip or knee arthroplasties. The indications for transfusions remain unclear despite published guidelines. Clinicians have adopted a restrictive transfusion threshold (75-80 g/L) but the consequences of such a strategy on functional outcome and quality of life are not known. First, we characterized the transfusion practice inside the Centre hospitalier de l’Université de Montréal (CHUM). Our hypothesis was that transfusion practice varies inside the CHUM due to uncertainty. A retrospective study of 701 charts of patients operated for a hip or knee arthroplasty was conducted. We observed that there was no difference among hospitals regarding the way transfusions are used and that physicians mainly based their decision to transfuse on a single variable, the hemoglobin concentration, adopting a restrictive transfusion strategy. Sixty-six percent of patients had a hemoglobin concentration under 100 g/L after surgery. Second, we evaluated the impact of this postoperative anemia on functional outcome and quality of life. We hypothesized that a threshold hemoglobin concentration exists below which these become impaired. A prospective, observational cohort study was conducted in 305 patients categorized in groups according to their postoperative hemoglobin concentration. Hemoglobin groups (≤ 80, 81-90, 91-100 and > 100 g/L) were similar in the evolution of the distance walked in six minutes, perception of effort, maximal dominant hand strength and quality of life scores. Thus, moderate anemia is not associated with an impaired functional recovery or quality of life early after hip and knee arthroplasties. Further studies will be required to determine the long-term consequences of a restrictive transfusion strategy in these patients.

Les accidents sont la cause la plus fréquente de décès chez l’enfant, la plupart du temps à cause d’un traumatisme cranio-cérébrale (TCC) sévère ou d’un choc hémorragique. Malgré cela, la prise en charge de ces patients est souvent basée sur la littérature adulte. Le mannitol et le salin hypertonique (3%) sont des traitements standards dans la gestion de l’hypertension intracrânienne, mais il existe très peu d’évidence sur leur utilité en pédiatrie. Nous avons entrepris une revue rétrospective des traumatismes crâniens sévères admis dans les sept dernières années, pour décrire l’utilisation de ces agents hyperosmolaires et leurs effets sur la pression intracrânienne. Nous avons établi que le salin hypertonique est plus fréquemment utilisé que le mannitol, qu’il ne semble pas y avoir de facteurs associés à l’utilisation de l’un ou l’autre, et que l’effet sur la pression intracrânienne est difficile à évaluer en raison de multiples co-interventions. Il faudra mettre en place un protocole de gestion du patient avec TCC sévère avant d’entreprendre des études prospectives. La transfusion sanguine est employée de façon courante dans la prise en charge du patient traumatisé. De nombreuses études soulignent les effets néfastes des transfusions sanguines suggérant des seuils transfusionnels plus restrictifs. Malgré cela, il n’y a pas de données sur les transfusions chez l’enfant atteint de traumatismes graves. Nous avons donc entrepris une analyse post-hoc d’une grosse étude prospective multicentrique sur les pratiques transfusionnelles des enfants traumatisés. Nous avons conclu que les enfants traumatisés sont transfusés de manière importante avant et après l’admission aux soins intensifs. Un jeune âge, un PELOD élevé et le recours à la ventilation mécanique sont des facteurs associés à recevoir une transfusion sanguine aux soins intensifs. Le facteur le plus prédicteur, demeure le fait de recevoir une transfusion avant l’admission aux soins, élément qui suggère probablement un saignement continu. Il demeure qu’une étude prospective spécifique des patients traumatisés doit être effectuée pour évaluer si une prise en charge basée sur un seuil transfusionnel restrictif serait sécuritaire dans cette population. / Trauma is the leading cause of death of children, with the burden of mortality related both to traumatic brain injury and hemorrhagic shock. Despite the frequency of trauma in the pediatric population, the management of these patients is often based on adult literature due the sparse amount of literature in pediatric trauma. The studies presented below were intended to establish current practice, and prepare for future prospective studies in pediatric trauma. The management of raised intracranial pressure (ICP) following traumatic brain injury (TBI) involves intracranial monitoring and the escalation of care to prevent secondary insults to the brain. Hyperosmolar therapy with mannitol (20%) and hypertonic saline (3%) are standard of care for the reduction of ICP, despite little evidence for their use. Our retrospective, single center study aimed to describe the clinical practice of hyperosmolar therapy in pediatric severe TBI, and its effect on ICP. We found that both mannitol and hypertonic saline are frequently used without a clear indication for one agent over another. There was insufficient power to confirm an effect on ICP, and multiple co-interventions given after boluses of hyperosmolar therapy may have contributed this lack of effect. In order to prospectively evaluate the effect of hyperosmolar therapy on ICP, a standardized approach to TBI care and hyperosmolar agents is necessary. Red blood cell transfusion is a key component of the management of the unstable trauma patient. Literature now suggests that transfusion is associated with increased mortality, and practices have shifted toward restrictive transfusion strategies in many clinical populations. We sought to describe the transfusion practices in pediatric trauma patients based on a secondary analysis of a large prospective study on blood loss in pediatric intensive care unit (PICU) patients. Compared to non-trauma patients, trauma patients were more likely to be transfused and transfused early in their course of stay. Younger age, higher PELOD and mechanical ventilation were associated with receiving a red blood cell transfusion in the PICU. Receiving a blood transfusion prior to PICU admission was most strongly associated with receiving a transfusion after PICU admission, suggesting ongoing bleeding in those transfused early. Future prospective studies geared specifically for trauma patients are necessary to determine whether osmotherapy for high ICP, and restrictive transfusion strategies can be applied to them, in order to improve the quality of the evidence based care provided to children.

Les transfusions sanguines sont fréquemment employées pour corriger l’anémie secondaire à une arthroplastie de la hanche ou du genou. Il n’y a cependant pas consensus sur les indications de transfuser. La tendance actuelle est d’utiliser une stratégie transfusionnelle restrictive (soit un seuil de 75-80 g/L d’hémoglobine) mais les conséquences d’une telle pratique sur la récupération fonctionnelle et la qualité de vie des patients sont mal connues. Dans un premier temps, nous avons caractérisé la pratique transfusionnelle au Centre hospitalier de l’Université de Montréal (CHUM). Notre hypothèse était que, devant l’imprécision des recommandations, la pratique transfusionnelle serait variable. Une étude rétrospective de 701 dossiers de patients ayant subi une arthroplastie de la hanche ou du genou a été réalisée. Nous avons observé que les transfusions étaient utilisées de la même façon dans les trois hôpitaux et que les médecins basaient leur décision de transfuser principalement sur un seul chiffre, la concentration d’hémoglobine, adoptant une stratégie restrictive. Soixante-six pourcent des patients avaient une concentration d’hémoglobine inférieure à 100 g/L au départ de l’hôpital. Dans un deuxième temps, nous avons évalué l’impact de cette anémie postopératoire sur la récupération fonctionnelle et la qualité de vie des patients. Notre hypothèse était qu’il existe une concentration d’hémoglobine en dessous de laquelle celles-ci sont atteintes. Une étude de cohorte prospective et observationnelle a été menée chez 305 patients regroupés selon leur concentration d’hémoglobine postopératoire. Les groupes d’hémoglobine (≤ 80, 81-90, 91-100 et > 100 g/L) étaient similaires dans l’évolution de la distance de marche en six minutes, de l’évaluation de l’effort fourni, de la force de préhension et des scores de qualité de vie. L’anémie modérée n’est donc pas associée à une atteinte de la récupération fonctionnelle et de la qualité de vie à court terme. D’autres études devront déterminer les conséquences à long terme d’une stratégie transfusionnelle restrictive sur ces patients. / Red blood cell transfusions are frequently used to treat anemia after total hip or knee arthroplasties. The indications for transfusions remain unclear despite published guidelines. Clinicians have adopted a restrictive transfusion threshold (75-80 g/L) but the consequences of such a strategy on functional outcome and quality of life are not known. First, we characterized the transfusion practice inside the Centre hospitalier de l’Université de Montréal (CHUM). Our hypothesis was that transfusion practice varies inside the CHUM due to uncertainty. A retrospective study of 701 charts of patients operated for a hip or knee arthroplasty was conducted. We observed that there was no difference among hospitals regarding the way transfusions are used and that physicians mainly based their decision to transfuse on a single variable, the hemoglobin concentration, adopting a restrictive transfusion strategy. Sixty-six percent of patients had a hemoglobin concentration under 100 g/L after surgery. Second, we evaluated the impact of this postoperative anemia on functional outcome and quality of life. We hypothesized that a threshold hemoglobin concentration exists below which these become impaired. A prospective, observational cohort study was conducted in 305 patients categorized in groups according to their postoperative hemoglobin concentration. Hemoglobin groups (≤ 80, 81-90, 91-100 and > 100 g/L) were similar in the evolution of the distance walked in six minutes, perception of effort, maximal dominant hand strength and quality of life scores. Thus, moderate anemia is not associated with an impaired functional recovery or quality of life early after hip and knee arthroplasties. Further studies will be required to determine the long-term consequences of a restrictive transfusion strategy in these patients.