The systematic use of evidence‐based methodologies and technologies enhances shared decision‐making in the 2018 International Consensus Conference on Patient Blood Management

Background and objectives Patient Blood Management (PBM) aims to optimize the care of patients who might need a blood transfusion. The International Consensus Conference on PBM (ICC‐PBM) aimed to develop evidence‐based recommendations on three topics: preoperative anaemia, red blood cell transfusion thresholds and implementation of PBM programmes. This paper reports how evidence‐based methodologies and technologies were used to enhance shared decision‐making in formulating recommendations during the ICC‐PBM. Materials & Methods Systematic reviews on 17 PICO (Population, Intervention, Comparison, Outcomes) questions were conducted by a Scientific Committee (22 international topic experts and one methodologist) according to GRADE (Grades of Recommendation, Assessment, Development and Evaluation) methodology. Evidence‐based recommendations were formulated using Consensus Development Conference methodology. Results We screened 17 607 references and included 145 studies. The overall certainty in the evidence of effect estimates was generally low or very low. During the ICC, plenary sessions (100–200 stakeholders from a range of clinical disciplines and community representatives) were followed by closed sessions where multidisciplinary decision‐making panels (>50 experts and patient organizations) formulated recommendations. Two chairs (content‐expert and methodologist) moderated each session and two rapporteurs documented the discussions. The Evidence‐to‐Decision template (GRADEpro software) was used as the central basis in the process of formulating recommendations. Conclusion This ICC‐PBM resulted in 10 clinical and 12 research recommendations supported by an international stakeholder group of experts in blood transfusion. Systematic, rigorous and transparent evidence‐based methodology in a formal consensus format should be the new standard to evaluate (cost‐) effectiveness of medical treatments, such as blood transfusion.


Appendix 4 List of continents and countries included in the participation list of the ICC-PBM 2018
Appendix 5 List of Institutions/Organizations that co-sponsored or contributed during the ICC-PBM 2018

Co-sponsors Contributors
The           Appendix 8

Appendix 6 Composition decision-making panels
Draft recommendations of the decisionmaking panels at the end of day 1 Definition and diagnosis of preoperative anaemia (PICO 1-2) • The panel recognizes that perioperative anaemia is an important risk factor for perioperative morbidity and mortality and therefore recommends to detect and classify anaemia early before major elective surgery (strong recommendation based on low certainty in the evidence of effects).
• The panel noticed that the thresholds for definition of anaemia are heterogeneous in the literature. Therefore, exact thresholds need to be addressed in future studies.

Treatment of preoperative anaemia (PICO 3)
• The panel decided not to recommend the use of prophylactic transfusion in adult perioperative elective surgery patients because there is no evidence of any advantage for this approach.
• The panel recommends using iron supplementation in adult preoperative elective surgery patients with iron deficiency anaemia to reduce RBC transfusion rate (strong recommendation based on low certainty in the evidence of effects). The panel recognizes that the current evidence is based only on studies published until 2015.
• The panel decided to formulate a recommendation for further research to address the effect of iron substitution on the subgroup of patients with preoperative iron deficiency.
• The panel decided to formulate no recommendation on the use of ESA monotherapy in adult preoperative elective cardiac surgery patients.
• The panel suggests not to use ESA therapy routinely in anaemic adult preoperative elective surgery patients (conditional recommendation based on low certainty in the evidence of effects).
• The panel suggests to use ESA therapy in addition to iron supplementation in adult preoperative elective major orthopaedic surgery patients with Hb levels <13 g/dl as desirable effect (reduced RBC transfusion rate) may outweigh potential undesirable effects for this subgroup of patients (conditional recommendation based on low certainty in the evidence of effects).
• The panel decided to formulate a recommendation for further research on the use of ESA+iron therapy in adult preoperative elective surgery patients with focus on long-term (un)desirable effects, optimal dose, type of surgery (particular in cancer surgery), co-presence of iron deficiency and cost-effectiveness.
The use of RBC transfusion triggers (PICO 4-PICO 14) • Critical care but clinically stable ICU (PICO 4-PICO 8): the panel drafted a strong recommendation in favour of using a RBC transfusion threshold of <7 g/ dl for the treatment of anaemia in critically ill adult patients who are not actively bleeding (strong recommendation, moderate level of evidence). This recommendation may not apply to patients with a history of coronary heart disease, other cardiovascular disease or brain injury. Patients with septic shock are part of this population (originally separate PICO 8) and the Hb <7 g/dl trigger represents the value used in the included trials.
• Orthopaedic surgery (PICO 5): the panel drafted a conditional recommendation in favour of using a RBC transfusion threshold of Hb <8 g/dl in patients with hip fracture with cardiovascular disease or risk factors (conditional recommendation, moderate level of evidence). This recommendation was justified by the fact that no effect on mortality (although wide 95% confidence interval) or functional outcomes (walk independently at 60 days) was present. However, uncertainty regarding undesirable effects such as acute myocardial infarction resulted in a conditional rather than a strong recommendation. The trigger of Hb <8 g/dl represents the value used in the included trials and major evidence gaps in the areas are still present. Therefore, an additional recommendation for further research was proposed.
• Non-cardiac surgery (PICO 5): the panel drafted a recommendation for further research on the use of restrictive transfusion triggers in non-cardiac surgery populations. A conditional recommendation for either strategy cannot be made because of the concern over the possibility for undesirable effects in the restrictive group.
• Acute gastrointestinal bleeding (PICO 6): the panel suggested to formulate a recommendation in favour of the use of a RBC transfusion threshold of Hb 7-8 g/dl in patients with acute gastrointestinal bleeding who are haemodynamically stable (conditional recommendation, low level of evidence). This proposal was justified by the evidence from the two included trials showing a lower mortality and a reduced RBC exposure and utilization in the restrictive transfusion group. It was noted by the panel that (1) the population in the PICO was defined as "acute gastrointestinal bleeding" whereas the study populations of the included studies were limited to acute "upper" gastrointestinal bleeding.
(2) No trials identified patients with lower gastrointestinal bleeding.
(3) Guidelines should emphasize that in the acutely bleeding patient, Hb is not the deciding factor for transfusion and (4) the included trials used Hb triggers (e.g. Hb <7 g/dl) to achieve specified Hb target ranges (e.g. Hb 7-9 g/dl).
• Coronary heart disease (PICO 7): the panel drafted a recommendation for further research on the use of restrictive transfusion triggers in adult patients with acute coronary syndrome or other ischaemic heart disease. This was justified by the overall low level of evidence and concern regarding undesirable effects on clinical outcomes (e.g. 30-day mortality) with a restrictive strategy.
• Cardiac surgery (PICO 9): the panel drafted a strong recommendation in favour of using a RBC transfusion threshold of <7Á5 g/dl in adult cardiac surgery patients (strong recommendation, moderate level of evidence). The trigger Hb <7Á5 g/dl represents the value used in the included trials, and no evidence of increased mortality or other undesirable effects was present together with a substantial reduction in RBC exposure and utilization.
• Haematology (PICO 10): no RBC transfusion trigger was recommended because of insufficient evidence (only two pilot studies in acute leukaemia (n = 149)) and no signal for undesirable effects. A recommendation for further research on RBC transfusion support in adult patients with haematological diseases (including non-malignant conditions such as haemoglobinopathies) was drafted.
• Oncology (PICO 11): no RBC transfusion trigger was recommended because no evidence was available. The panel decided that the only available trial was conducted in a postoperative surgical oncology setting in an intensive care unit and was therefore classified under PICO 5 (non-cardiac surgery).
• Neurology -Central nervous system injuries (PICO 12): no RBC transfusion trigger was recommended due to a very low level of evidence for all (critical) outcomes. The panel decided to formulate a recommendation for further research on the use of restrictive transfusion triggers in patients with CNS injury.
• Neurology -Cerebral perfusion disorders (PICO 13): no RBC transfusion trigger was recommended because no evidence for any outcome related to a restrictive transfusion strategy was available. Indeed, the included trial randomised patients to a Hb trigger of 10 g/dl (intervention group) or 11.5 g/dl (control group). The intervention group was not considered as a restrictive transfusion strategy group.
• Acute bleeding (PICO 14): because only evidence from one pseudo-randomized trial from 1956 was available, no RBC transfusion trigger was recommended. The panel view was that a Hb concentration alone should not be used to determine the need for transfusion in an acutely bleeding scenario (i.e. major haemorrhage) Implementation of PBM programmes  • The panel drafted a recommendation in favour of using comprehensive PBM programmes to improve appropriate RBC utilization (conditional recommendation based on low certainty in the evidence of effects).
• The panel drafted a recommendation in favour of using behavioural interventions (transfusion guideline/audit/form/education) to improve appropriate RBC utilization (conditional recommendation based on very low certainty in the evidence of effects).
• The panel drafted a recommendation in favour of using electronic/computerized decision support systems to improve appropriate RBC utilization (conditional recommendation based on low certainty in the evidence of effects).
• General research priorities in the field of PBM implementation were formulated including (1) measuring the impact on FFP/PLT/cryo utilization and clinical outcomes, (2) designing well-conducted observational studies (e.g. time interrupted series), (3) the assessment of compliance data, (4) measuring the cost-effectiveness of PBM programmes and the relative effectiveness of different types of decision support systems.