Feature Articles Survey on transfusion practices of pediatric intensivists* Caroline Laverdière, MD, FRCPC; France Gauvin, MD, FRCPC, FAAP; Paul C. Hébert, MD, FRCPC, MHSc; Claire Infante-Rivard, MD, FRCPC, PhD; Heather Hume, MD, FRCPC; Baruch J. Toledano, MD, FRCPC, FAAP; Marie-Claude Guertin, PhD; Jacques Lacroix, MD, FRCPC, FAAP; for the Canadian Critical Care Trials Group A significant proportion of criti- cally ill children receive red blood cell (RBC) transfusions during their pediatric intensive care unit (PICU) stay (15% in the PICU of Sainte-Justine Hospital) (1). There are few data on the indications for packed RBC transfusions in critically ill children, and therefore one may ask on what evidence pediatric intensivists base their transfusion decisions. The cost/efficacy of RBC transfu- sion is not clear: The risk of infection from an RBC transfusion is well recognized, but some data suggest also that transfusion may be a risk factor for nosocomial infec- tion and multiple organ dysfunction syn- drome (2). Moreover, a number of articles report that RBC transfusions are not opti- mally used (2). A survey reported a signifi- cant variation in transfusion practices among Canadian intensivists who take care of critically ill adults (3). The practice pat- tern that is advocated by pediatric intensiv- ists with respect to RBC transfusion has not been studied. Thus, we undertook a survey among pediatric intensivists to find out their stated practice patterns in regard to RBC transfusion and to determine whether a consensus existed regarding the indica- tions for RBC transfusion therapy. METHODS Study Design. We undertook a cross- sectional, self-administered survey designed to detect determinants of RBC transfusion in critically ill children. The questions were de- veloped to find out if the characteristics of the respondents, patients, and/or PICUs changed the threshold hemoglobin (Hb) concentration at which pediatric intensivists transfused RBCs. The questionnaire was scenario-based, and the possible clinical determinants were changed one at a time in each of the four clinical scenarios. The survey aimed to elicit information about the physicians’ behavior in clinical practice (4). Study Population. The questionnaire was mailed to English- and French-speaking in- tensivists practicing in academic PICUs in Canada and French-speaking European coun- tries (mostly France, Belgium, and Switzer- land). The mailing list of Canadian pediatric intensivists was based on the list of members of the Canadian Critical Care Society and the Canadian Pediatric Society; we also phoned pediatric critical care program directors to ensure that our listing was complete. The mailing list of French-speaking pediatric in- tensivists was constructed from the member- ship of the Groupe francophone de Réanima- tion et d’Urgence pédiatrique and the Société de Réanimation de Langue française. We de- cided a priori to exclude retired members, physicians who have ceased to work in a PICU, intensivists working outside the jurisdictions under study, neonatologists, intensivists working with adults, and physicians who had outdated addresses. *Also see p. 381. From the Hematology Division (CL, HH) and Divi- sion of Pediatric Critical Care (FG, BJT, JL), Depart- ment of Pediatrics, Sainte-Justine Hospital, Université de Montréal, Montréal, Québec, Canada; Critical Care Program (PCH), Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada; Joint Departments of Epidemiology and Biostatistics (CI-R), and Occupa- tional Health, Faculty of Medicine, McGill University, Montréal; and Research Center (M-CG), Institut de Cardiologie, Université de Montréal. Supported, in part, by the Canadian Institutes of Health Research and the Fondation de l’Hôpital Sainte- Justine. Presented, in part, at the 11th Colloquium on Pedi- atric Intensive Care Medicine, Chicago, September 1998. Address requests for reprints to: Jacques Lacroix, MD, FRCPC, FAAP, Sainte-Justine Hospital, 3175 Côte Sainte-Catherine, Montreal (Quebec), Canada H3T 1C5. E-mail: jacques_lacroix@ssss.gouv.qc.ca Copyright © 2002 by the Society of Critical Care Medicine and the World Federation of Pediatric Inten- sive and Critical Care Societies DOI: 10.1097/01.PCC.0000031371.88694.2B Objective: To describe the red blood cell transfusion practices of pediatric intensivists. Design: Cross-sectional self-administered survey. Setting: Pediatric intensive care units. Patients: Academic pediatric intensivists. Interventions: None. Measurements and Main Results: Scenario-based survey among English- or French-speaking intensivists from Canada, France, Belgium, or Switzerland, working in tertiary-care pediat- ric intensive care units. Respondents were asked to report their decisions regarding transfusion practice with respect to four scenarios: cases of bronchiolitis, septic shock, trauma, and the postoperative care of a patient with Fallot’s tetrad. The response rate was 71% (163 of 230). The overall baseline hemoglobin transfusion threshold that would have prompted intensivists to transfuse a patient ranged from 7 to 13 g/dL (70 –130 g/L) within almost all scenarios. There was a significant difference between scenarios of the average baseline hemoglobin transfusion thresh- olds (p < .0001). A low PaO 2 , a high blood lactate concentration, a high Pediatric Risk of Mortality score, active gastric bleeding, emergency surgery, and age (2 wks) were important determinants of red blood cell transfusion, whereas none of the respondents’ personal characteristics were. The average volume of packed red blood cells transfused in the four scenarios did not differ signif- icantly. Conclusions: This survey documented a significant variation in transfusion practice patterns among pediatric critical care prac- titioners with respect to the threshold hemoglobin concentration for red blood cell transfusion. The volume of packed red blood cells given was not adjusted to the hemoglobin concentration. (Pediatr Crit Care Med 2002; 3:335–340) KEY WORDS: blood; child; critical care; erythrocyte; hematocrit; hemoglobin; intensive care; pediatrics; physician practice pattern; practice; risk factors; surveys; transfusion. 335 Pediatr Crit Care Med 2002 Vol. 3, No. 4