CARDIOVASCULAR JOURNAL OF AFRICA • Volume 34, No 3, July/August 2023 AFRICA 165 For the HBM programme, first, transfusion characteristics in the emergency service, intensive care unit (ICU) and operating room were determined. Data were recorded at hours when transfusion requests were heaviest. The use of autologous blood, habits of transfusion of fresh frozen plasma (FFP), fresh whole blood (FWB) and other blood components, and donation of pre-operative blood components were tabulated. Proposals for solutions to problems identified were addressed through shared wisdom. In January 2019, a model of transfusion processes was developed as a quality-improvement project. The data collected before and after the start of the programme were compared. We first reviewed transfusion-related workflow step by step to detect problems in the process (Fig. 1).11,12 We created an acceptable standardised transfusion threshold, ranging from a more restrictive to a more liberal threshold, depending on the patient populations. The core team observed physicians’ transfusion practice and decision making. The pre-operative quantity of blood components prepared were three units of packed red blood cells (PRBC), two units of FWB and two units of FFP. For every FWB and PRBC unit ordered, a cross-match was made. Two units of FFP were routinely administered concomitantly with protamine at the end of cardiopulmonary bypass (CPB). FFP was the most-often requested blood component in the emergency department (ED). Patients with international normalised ratio (INR) values of > 4.5 received FFP transfusion routinely because of the high risk of bleeding. Cryoprecipitate transfusions were administered without measuring fibrinogen values, except with massive transfusions. Platelets were ordered for postoperative patients who had continued bleeding, and platelets were routinely administered to patients with platelet counts of < 100 000 cells/mm3. Peri-operative autologous blood collection was rare. Use of cell recovery (Cell Saver®) was limited. One-third of all bloodcomponent orders was made during the 17:00 shift change of the ICU. Initially, all surgeons, anaesthesiologists and internal medicine physicians, then perfusionists, anaesthesiology technicians and nurses were given training on the risks, benefits and alternatives of blood transfusion. The training modules were tailored to the transfusion steps for which each professional was responsible. Blood component usage, approach to transfusion reactions, transfusion policies of our centre and our plans were explained, with examples taken from our daily clinical practice. Reminders were provided through e-mail and hospital-informed pop-up windows. Use of FWB and FFP transfusions was high and for numerous indications, therefore, emphasis was placed on reducing the number of FWB and FFP transfusions. Two units of PRBC were prepared for pre-operative patients. Also, orders for FFP were not permitted before operations. FWB orders were limited to situations where FWB was felt necessary, such as massive transfusions or high INR due to warfarin usage or liver failure. Clinicians determined the PRBC transfusion threshold to be a haemoglobin (Hb) value of 9 g/dl. Physicians were encouraged to evaluate the patient’s haemodynamic status, electrocardiogram, and arterial blood gas values before giving transfusions, rather than transfusing based on Hb value alone. Teams responsible for elective operations and invasive procedures were asked to reschedule if their patients had INR values > 1.8. Clinicians caring for extracorporeal membrane oxygenation, intra-aortic balloon pump and left ventricular assist devices were allowed more liberal transfusion. FFP given for high INR was most frequently given in the ED. The use of FFP for high INR is common in our institution, but it is not recommended. A flow chart for FFP was prepared, based on current guidelines.8,13-16 Patients with bleeding risk or active bleeding were transferred to the ICU before transfusion unless the bleeding was critical. The use of peri-operative autologous blood by anaesthesiologists was increased through training and practice. Autologous blood was stored in the operating rooms at room temperature for up to six hours. The practice of routinely transfusing two units of FFP immediately after protamine injection at the end of CPB was discontinued. Cryoprecipitate was transfused as part of massive transfusions and/or bleeding with fibrinogen values < 100 mg/dl. At the end of evening (17:00) rounds, a senior physician reviewed the blood transfusions given to haemodynamically stable patients, and the staff was given feedback when inappropriate transfusions occurred. To aid in evaluating the transfusion practices, a hospital blood-management follow-up form was constructed after receiving input from all team members (Fig. 2). With this form and the patients’ demographic data, a practical guide was created for decision making on the transfusion of blood components. The form was adapted to fit the features of our centre and was easy to understand and complete. Statistical analysis IBM SPSS 15.0 was used in the statistical analysis. The frequency distribution for each variable was compared between the two years. The chi-squared test with continuity correction was performed. A p-value < 0.05 was considered statistically significant. Identifying problems in the field A core team including cardiovascular surgeon, anesthesiologist and critical care doctor, physician in charge of transfusion and haemovigilance nurses was set up that was responsible for holding presentations, training, data monitoring and auditing of the programme. A hospital blood management follow-up form was created after receiving information from all team members to assist in the evaluation of transfusion practices. By controlling the transfusion forms, bedside transfusion follow up was started with daily monitoring, evaluation and feedback. In this step, attention was paid to the indication of the product used, the duration of administration and the suitability of the current clinical and laboratory parameters. Finally, the obtained data were analysed to compare before and after hospital blood management practice. Fig. 1. Stages of the transfusion programme model
RkJQdWJsaXNoZXIy NDIzNzc=