Patient blood management (PBM) or, more precisely, patient-focused blood management, ‘is the timely application of evidence-based medical and surgical concepts designed to maintain haemoglobin (Hb) concentration, optimise haemostasis and minimise blood loss in an effort to improve patient outcome’ (www.sabm.org). By definition, the approach is multidisciplinary and focuses on the treatment of each individual patient in whom significant blood losses are likely to occur and where transfusion of blood products is part of the established treatment.1 PBM is known best in the management of elective surgery and should be applied to any procedure likely to result in excessive bleeding, postprocedural anaemia and the use of blood products. The ‘three pillars’ of PBM require that erythropoiesis be optimised, blood losses minimised and anaemia managed appropriately.2 Finally, PBM is a continuous process, initiated early in the preoperative period and continued intra and postoperatively.
High variability in transfusion practices has been demonstrated in studies comparing blood usage between institutions and/or countries. Improvement of existing practice may be achieved through a benchmarking process that consists of a structured continuous collaborative procedure in which comparisons for selected indicators are used to identify factors that, when implemented, will increase performance.3 Despite an important number of publications describing transfusion practices in different countries, few were true benchmarking studies and none originated from Europe. European healthcare providers feel a pressing need to integrate PBM in the routine management of elective surgical patients to improve transfusion practice.2
NATA, formerly the Network for the Advancement of Transfusion Alternatives, now the ‘Network for the advancement of patient blood management, haemostasis and thrombosis’, is a multidisciplinary society comprising anaesthesiologists, intensivists, surgeons, haematologists, blood bankers and transfusion medicine specialists interested in promoting best practices in the fields of PBM, haemostasis and thrombosis. NATA initiated a benchmark project across 11 international centres with the aim of providing the basis for educational strategies to implement optimal usage of blood and blood products in Europe and Canada.
The project was designed as a prospective, observational, multicentre study with online data collection. Centres interested in participating were selected on a voluntary basis among suggestions made by members of the NATA scientific committee. The study was conducted in 11 centres from January 2010 to June 2011. No informed consent and no ethical approval were required for this type of observational study because of its strictly anonymous design. However, ethics committee approval was obtained if required locally.
Three standard elective surgical procedures were investigated: primary unilateral total hip replacement (THR) using either cemented or noncemented implants, unilateral knee replacement (TKR) and coronary artery bypass graft (CABG) surgery. Reoperations were included for all procedures. Patients less than 18 years of age, emergency surgery and patients with hereditary coagulation disorders were excluded.
The study period per centre was 6 months. The expected sample size per indication varied between 50 and 150 depending on the workload of each centre. When the specified sample size was reached or the recording period elapsed, whichever occurred first, the centres contacted the study management team, which then decided whether recording could stop or had to be continued for a defined period. The overall aim was to include a minimum of 1650 patients.
A web-based electronic data capture system for data acquisition was provided by the Austrian Institute of Technology GmbH (Graz, Austria). Detailed explanations and instructions were provided to the study participants in a printed manual and on the website. Data were recorded directly into the study database via a dedicated website. The system provided login names and passwords dedicated for registration of patients, monitoring of recruiting progress, query management and source data verification. Automatic data entry plausibility checks and mandatory data items enforced high data quality. Four of the investigators (HG, JFH, AM, PVDL) conducted one on-site monitoring visit at the end of recruitment. This visit included checking the inclusion/exclusion criteria, checking compliance with the consecutive recruitment process, source data verification (random samples) and on-site query management. Findings of the monitoring visit were reported to the study management team.
Data on patients’ characteristics, surgical procedure, blood-sparing interventions, Hb concentrations and blood product transfusions were collected. A detailed description of all data collected is presented as supplemental digital content, http://links.lww.com/EJA/A100. Our primary outcome measures were the incidence and volume of red blood cells (RBC) transfused according to the procedure. A multivariate logistic regression analysis was used to identify variables independently associated with RBC transfusions.
Categorical variables are presented as absolute and relative frequencies, and numerical variables as mean ± SD. Differences between centres and, wherever appropriate, between groups of patients who were or who were not transfused were analysed with χ2 tests and log-linear models (categorical variables) and by one or two-way analysis of variance (numerical variables). As this is an observational study, P values are given for information only. Given the nature of the study, no sample size calculation was deemed necessary.
Variables associated with RBC transfusion were analysed with logistic regression models, including age, sex, BMI, American Society of Anesthesiologists’ physical status, preoperative and lowest postoperative Hb concentrations, duration of surgery, use of washed and unwashed shed blood, use of platelet aggregation inhibitors and type of anaesthetic. In CABG patients, EuroSCORE, number of grafts, administration of tranexamic acid and use of extracorporeal circulation were used as additional independent variables.
Hb concentrations were converted to percentages of the WHO cutoff values for anaemia (men 13 g dl−1, women 12 g dl−1) before the analysis, as described elsewhere,4 to allow elimination of the effect of sex and combination of men and women in one analysis. Similarly, to adjust for baseline differences in total RBC volume and for sex differences, lost and transfused RBC volumes were also converted to percentage of the preoperative circulating RBC volume.
Ten European centres and one Canadian centre participated in the NATA benchmark project (Table 1). One centre did not contribute THR or TKR data whereas four centres did not contribute CABG data. The total number of patients registered in the study was 2470. The number of complete records totalled 2431; calculation of lost volume of RBC was not possible in 144 records in which postoperative Hb values were missing (Fig. 1).
Patient characteristics, evolution of Hb concentrations and blood losses for all patients are presented in Table 2. The following variables were significantly different between centres (data not shown):
- For THR: age, BMI, duration of surgery, preoperative Hb, postoperative day (POD) 5 Hb, minimum postoperative Hb, lost volume of RBC, use of platelet inhibitors, use of tranexamic acid, unwashed blood reinfusion, type of anaesthesia and incidence of preoperative anaemia.
- For TKR: age, BMI, duration of surgery, preoperative Hb, POD 5 Hb, minimum postoperative Hb, lost volume of RBC, use of platelet inhibitors, use of tranexamic acid, use of cell salvage, unwashed blood reinfusion, type of anaesthesia and incidence of preoperative anaemia.
- For CABG: age, duration of surgery, EuroSCORE, number of grafts, duration of extracorporeal circulation, preoperative Hb, POD 5 Hb, minimum postoperative Hb, lost volume of RBC, use of platelet inhibitors, use of tranexamic acid, use of cell salvage, unwashed blood reinfusion and incidence of preoperative anaemia.
The incidences of preoperative anaemia and of its management with iron and/or erythropoietin were significantly different between centres and for the three procedures (Table 3). In THR patients, the incidence of preoperative anaemia ranged from 4.7 to 30.4% whereas treatment of anaemia ranged from 0 to 40%. In TKR patients, the incidence of preoperative anaemia ranged from 2.9 to 40.0% whereas treatment of anaemia ranged from 0 to 41%. In CABG patients, the incidence of preoperative anaemia ranged from 17.7 to 40.0% whereas treatment of anaemia ranged from 0 to 15%. Two sites were using only iron, one only erythropoietin, six both and two neither of these therapeutic modalities. Overall, there was no correlation between the incidence of preoperative anaemia and the frequency of its management (i.e. centres with the higher incidence of preoperative anaemia were not those where anaemia was most frequently treated).
The incidence of tranexamic acid and cell salvage use was significantly different between centres for the three procedures (Table 2). Use of preoperative autologous donation was less than 1.5% in THR and TKR. The incidence of use of different anaesthetic techniques (general versus locoregional) was significantly different between centres for both THR and TKR.
The incidence of transfusion differed significantly between centres for THR (range 7 to 95%), TKR (range 3 to 100%) and CABG (range 20 to 95%). The volume of RBC transfused was significantly different between centres for THR (range 12 to 39% of preoperative circulating RBC volume) and CABG (range 25 to 52% of preoperative circulating RBC volume) (Table 4). The incidences of fresh frozen plasma and platelet concentrate transfusion were also significantly different between centres for CABG.
Variables independently associated with RBC transfusion in patients undergoing THR and TKR, as identified by multivariate analysis (logistic regression), were the minimum postoperative Hb concentration, the preoperative Hb concentration, lost volume of RBC, general anaesthesia and female sex. In CABG patients, the independently associated variables were the preoperative Hb concentration, lost volume of RBC, female sex, cell saver use and the administration of tranexamic acid. The identified variables correctly classified patients in over 90% of cases for all three procedures (Table 5).
The relationship between RBC transfusions, preoperative anaemia, minimum postoperative Hb concentration and RBC losses varied considerably between centres for the three procedures (Figs. 2–4). The figures also suggest that this relationship varied markedly within each centre according to the different surgical procedures that were performed.
The NATA benchmark project reviewed the charts of 2431 patients undergoing major hip, knee and CABG surgery in 11 centres interested in developing PBM. Overall, there was a considerable difference in the incidence of transfusion and of the volume of RBC transfused between centres for the three procedures. Patient characteristics (including the incidence of preoperative anaemia) and their overall perioperative management were also considerably different between centres for the three procedures. For all three procedures, preoperative Hb concentration, lost RBC volume and female sex were independently associated with RBC transfusion. Minimum postoperative Hb concentration was also independently associated with RBC transfusion for hip and knee surgery.
Taken together, these observations highlight two major elements. First, implementation of PBM remains extremely variable between centres despite major educational efforts aimed at disseminating knowledge in this area. Thus, there is ample room for improvement of PBM, despite the fact that the contributing secondary and tertiary care centres were already interested in PBM and in participating in our benchmarking process. Second, the development of PBM strategies should take into account not only the individual needs of patients but also the specificities of the individual healthcare facilities.
The variables independently associated with RBC transfusion clearly indicate that patient characteristics and therapeutic options chosen by the different care providers explain the important variability observed in transfusion practice among centres and surgical procedures. Although difficult to evaluate, the role of healthcare facilities in the different countries cannot be neglected. Considerations associated with healthcare facilities comprise not only the accessibility to various alternative techniques such as intravenous iron, erythropoietin, antifibrinolytic agents and perioperative cell salvage but also to their costs, which vary markedly among countries. Another factor that may have contributed to the observed variability is the relative contribution of the anaesthesiologists to the pre and postoperative care of the patients, which also varied significantly among and within centres according to the different surgical procedures.
PBM relies on three main strategies to achieve its goals: optimise the patient's RBC mass, minimise perioperative blood losses and optimise the patient's tolerance to anaemia.2
Optimising the patient's RBC mass requires the detection of preoperative anaemia and delaying elective surgery until the underlying disorders can be diagnosed and corrected if possible. At present, preoperative anaemia is often ignored, surgery proceeds as planned and RBCs are transfused when deemed necessary.4 However, cohort studies (in cardiac, noncardiac and vascular surgery) have shown that preoperative anaemia is associated with an increased incidences of postoperative adverse events and mortality.5–7 In addition, preoperative anaemia and a low intraoperative Hb concentration are important predictors of perioperative allogeneic RBC transfusions. Blood transfusions entail a number of known (e.g. transmission of infectious agents, transfusion reactions, medical errors, ABO mismatch, transfusion-related acute injury, transfusion-associated circulatory overload, etc.) and less well known risks. The latter are related, in essence, to emergent pathogens and to immunomodulation (increased incidence of infections, cancer recurrence).8 At present, it remains difficult to determine whether the observed adverse events and mortality are related to preoperative anaemia per se or to the increase in allogeneic transfusions secondary to anaemia. Nevertheless, the overall outcome is negative so preoperative anaemia must be managed in a timely fashion. Unfortunately, management of preoperative anaemia remains inadequate9,10 even in centres where audits are conducted on a regular basis.11
Local healthcare organisation must be adapted to optimise preoperative patient management. In practice, patients must be seen sufficiently ahead of time (at least 28 days) by clinicians well aware of PBM principles. The essential clinical, laboratory and pharmacological resources must be allotted to the management of the anaemic patient scheduled for surgery. Although variability in the incidence of preoperative anaemia among centres may be related to different factors (related to the population's characteristics, but also to the moment the Hb concentration was measured preoperatively), our results confirm that adherence to the principles of the first pillar of PBM needs to be improved.2,4,11 High-quality evidence in that direction would definitely improve adherence to PBM principles.
Minimising perioperative blood losses entails meticulous patient preparation and management to detect and manage acquired (most often pharmacological) or congenital bleeding disorders. A close collaboration between anaesthesiologists, haematologists, nurses, perfusionists and surgeons is mandatory to optimise the team's blood conservation approach in the perioperative period. The surgeon's contribution in this area (meticulous haemostatic and surgical techniques) is paramount, although often underestimated. Our results suggest that anaesthetic technique may also play a significant role in blood conservation in major orthopaedic surgery.
Interestingly, comparing our results with those of the Orthopedic Surgery Transfusion Hemoglobin European Overview (OSTHEO) study12 for hip and knee surgery, a significant evolution in practices can be observed. Overall, the incidence of anaemia is lower, and preoperative autologous donation has almost disappeared. Also, the use of tranexamic acid has increased dramatically, which may help to explain the lower calculated RBC blood losses (Table 2) reported in our study.
Surprisingly, in CABG surgery, the use of cell salvage and the administration of tranexamic acid were independently associated with increased perioperative transfusion. This may be explained by their use in procedures with a high risk of bleeding.
The third pillar of PBM relies essentially on the use of restrictive, evidence-based transfusion strategies. Several articles and meta-analyses have been published recently on the use of restrictive transfusion strategies in different patient populations. Overall, restrictive transfusion strategies do not appear to impact on the rate of adverse events (mortality, cardiac events, myocardial infarction, stroke, pneumonia and thromboembolism) compared with liberal transfusion strategies.13 In major orthopaedic surgery, there is strong evidence that adoption of a restrictive RBC transfusion strategy is well tolerated, even in elderly patients.14,15 In cardiac surgery, the evidence is not as clear, with studies reporting conflicting results.16,17 Our results did not allow us to determine the Hb concentration used to trigger RBC transfusion in the different centres for the three different operations. However, they clearly indicate a marked variability in transfusion practice among centres. Increased physician awareness and education, with the assistance of real-time clinical support, has been shown to effectively decrease RBC transfusions18 and could help decrease the observed variability.
We found that female sex was an independent predictor of perioperative RBC transfusion in all three surgical procedures. These results were observed despite adjustments made to eliminate the effect of sex on total RBC volume and preoperative Hb concentration. This is in contrast with other recent studies that used a similar methodology.4,11 Nevertheless, our observation underlines the importance of meticulous PBM in female patients.
Interpretation of the results of our study should take into account a number of limitations and strengths. The duration of time between data collection (2010 to 2011) and reporting of the results was longer than usual. At least two reasons explain the delay. First, PVDL conducted most on-site monitoring visits at the end of recruitment, which took some time before the database could be locked. Second, and most importantly, we waited more than 1 year for the results of the statistical analysis of the study, assigned to an independent statistician who developed a serious illness. Sadly, Professor Peter H. Rehak died prior to completion of the analyses which, subsequently, we had to complete. Although our data may not reflect the situation in 2016, it is interesting to note that the recent, large PBM study by Meybohm et al.19 reported an ‘increased’ incidence of anaemia on admission (36.2%) after the systematic implementation of a PBM programme in four German University Hospitals. Several centres surveyed by NATA reported an incidence of anaemia on admission lower than that 5 years earlier (Table 3).
As for all prospective observational studies, the risk of bias cannot be ruled out and deducing causation from the observed correlations is not possible. This study was not designed to assess the impact of various measures on blood conservation, contrary to the study by Meybohm et al.19 In addition, our findings cannot be generalised because only centres with an interest in PBM were included. Further, we compared practice in 10 European centres and in one Canadian centre, which may not be optimal because current practices across centres around the world will differ, as PBM is progressively implemented in various countries. Nonetheless, no centre stood out, either positively or negatively, as being different from the rest of the group.
We evaluated transfusion practices in three surgical procedures that are known to consume a substantial number of blood products in 11 different high development index countries. The results reported confidentially to each centre will allow them to implement tailored measures to improve their PBM strategies. Evaluation of progress will be the second step of the NATA benchmark project. According to Clevenger et al.20 ‘Strategic adoption of patient blood management in surgical practice is recommended, and will reduce costs and improve outcomes in surgery’.
In conclusion, our results demonstrate marked variability between centres in the management of RBC transfusion. The relative importance of the factors identified to explain RBC transfusion will differ across institutions, some factors being related to the patient whereas others are related to the health facility. Consequently, the development of PBM strategies should take into account not only the individual needs of patients but also the specificities of the individual healthcare facilities.
Acknowledgements relating to this article
Assistance with the study: the authors are extremely grateful to the following collaborators for their generous contribution to the NATA benchmark Study: Drs E. Bisbe, T. Villar Colmenero, L. Abarca Vilchez, Barcelona, Spain; Dr D. Filipescu, Bucharest, Romania; Dr H. Gombotz, Linz, Austria; Dr U. Kaizers, Leipzig, Germany; Drs M. Lalemand and D. Datoussaid, Brussels, Belgium; Dr M. Lancé, Maastricht, The Netherlands; Drs A. Maniatis, S. Siminelakis, M. Koutentakis, E. Zervou and Mrs L. Dova, O. Valari, Greece; Drs A. Noordgaard, K. Jepsen, O. Jans, Copenhaghen, Denmark; Dr P. Radtke, London, the United Kingdom; Dr N. Rosencher, Paris, France; Mrs Monique Ruel, Montréal, Canada. We would also like to acknowledge the participation of Mr François Christory, executive director of NATA, and the late Professor Peter H. Rehak for his contribution to the statistical analysis of the data.
Financial support and sponsorship: the NATA benchmark project was sponsored entirely by the Network for the advancement of patient blood management, haemostasis and thrombosis. No commercial interests were involved.
Conflicts of interest: none.
Presentation: preliminary results were presented at the 12th Annual Meeting of NATA held in Dublin, April 2011.
1. Gombotz H. Patient blood management: a patient-orientated approach to blood replacement with the goal of reducing anemia, blood loss and the need for blood transfusion in elective surgery. Transfus Med Hemother
2. Shander A, Van Aken H, Colomina MJ, et al. Patient blood management in Europe. Br J Anaesth
3. Apelseth TO, Molnar L, Arnold E, et al. Benchmarking: applications to transfusion medicine. Transfus Med Rev
4. Gombotz H, Rehak PH, Shander A, et al. Blood use in elective surgery: the Austrian benchmark study. Transfusion
5. Ranucci M, Di Dedda U, Castelvecchio S, et al. Impact of preoperative anemia on outcome in adult cardiac surgery: a propensity-matched analysis. Ann Thorac Surg
6. Wu WC, Schifftner TL, Henderson WG, et al. Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA
7. Gupta PK, Sundaram A, Mactaggart JN, et al. Preoperative anemia is an independent predictor of postoperative mortality and adverse cardiac events in elderly patients undergoing elective vascular operations. Ann Surg
8. Goodnough LT, Shander A. Patient blood management. Anesthesiology
9. Lasocki S, Krauspe R, von Heymann C, et al. PREPARE: the prevalence of perioperative anaemia and need for patient blood management in elective orthopaedic surgery: a multicentre, observational study. Eur J Anaesthesiol
10. Gross I, Seifert B, Hofmann A, et al. Patient blood management in cardiac surgery results in fewer transfusions and better outcome. Transfusion
11. Gombotz H, Rehak PH, Shander A, et al. The second Austrian benchmark study for blood use in elective surgery: results and practice change. Transfusion
12. Rosencher N, Kerkkamp HE, Macheras G, et al. Orthopedic Surgery Transfusion Hemoglobin European Overview (OSTHEO) study: blood management in elective knee and hip arthroplasty in Europe. Transfusion
13. Carson JL, Carless PA, Hebert PC. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev
14. Carson JL, Sieber F, Cook DR, et al. Liberal versus restrictive blood transfusion strategy: 3-year survival and cause of death results from the FOCUS randomised controlled trial. Lancet
15. Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med
16. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA
17. Murphy GJ, Pike K, Rogers CA, et al. Liberal or restrictive transfusion after cardiac surgery. N Engl J Med
18. Goodnough LT, Shieh L, Hadhazy E, et al. Improved blood utilization using real-time clinical decision support. Transfusion
19. Meybohm P, Herrmann E, Steinbicker AU, et al. Patient blood management is associated with a substantial reduction of red blood cell utilization and safe for patient's outcome: a prospective, multicenter cohort study with a noninferiority design. Ann Surg
2016; doi: 10.1097/SLA.0000000000001747. [Epub ahead of print].
20. Clevenger B, Mallett SV, Klein AA, et al. Patient blood management to reduce surgical risk. Br J Surg