The administration of blood components: a British Society for Haematology Guideline.

Abstract

This guideline was compiled according to the BSH process at http://www.b-s-h.org.uk/guidelines/. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) nomenclature was used to evaluate the levels of evidence and to assess the strength of recommendations. The GRADE criteria can be found at http://www.gradeworkinggroup.org. This guideline represents the minimum requirements for the safe administration of blood component transfusions. These are intended to provide the foundations for organisations to build on when developing their own local policies and guidelines. However, it should be recognised that the more complex the procedures are, the more open to error they become. A search of the published literature was undertaken using MEDLINE, EMBASE, CINAHL, the Cochrane Library and the Internet using search terms (and relevant MESH terms) blood, transfusion, administration, documentation, patient, identification, safety, error, consent, authorisation, prescription, electronic, information technology, bar code, duration, 30 min, 4 h, bacteria, monitoring and observations. This search covered the period from January 2009 (since the last version of this guideline) to January 2015 and was limited to English language. There is very little high-quality evidence to prove the efficacy of many of the specific interventions and procedures to improve the safety of the administration of blood components, and so professional experience and expert opinion of the writing group were also employed. The review of the manuscript was performed by the British Society for Haematology (BSH) Guidelines Committee Transfusion Task Force, the BSH Guidelines Committee and the transfusion sounding board of BSH. It was also on the member section of the BSH website for comment. It has also been reviewed by the members of the British Blood Transfusion Society (BBTS), including the BBTS Transfusion Practitioner Special Interest Group, the Serious Hazards of Transfusion (SHOT) and the National Blood Transfusion Committee (NBTC), including the NBTC Patient Involvement Working Group and the Royal College of Midwives. These organisations do not necessarily approve or endorse the contents. Errors in the requesting, blood sampling, collection and administration of blood components (red blood cells, platelets and plasma components) lead to significant risks for patients. Since its launch in 1996, the Serious Hazards of Transfusion (SHOT) scheme has continually shown that ‘incorrect blood component transfused’ episodes are a frequently reported transfusion hazard. These wrong blood incidents are mainly due to human error leading to misidentification of the patient and can lead to life-threatening haemolytic transfusion reactions and other significant morbidities. The National Comparative Audits (NCA) of England and Wales (2003, 2005, 2009, 2011) have shown that patients are placed at risk of avoidable complications of transfusion through misidentification and inadequate monitoring. This guidance concerns transfusion of all blood components (red blood cells, platelets, fresh-frozen plasma, cryoprecipitate and granulocytes), regardless of whether patients are being transfused in hospital or in the community. This guidance does not relate to autologous cell-salvaged blood, for which additional aspects need to be considered (guidance is available from the UK Cell Salvage Action Group (UKCSAG) at http://www.transfusionguidelines.org/transfusion-practice/uk-cell-salvage-action-group). A key action point summary of this guideline is available in Fig. 1. For additional neonatal or paediatric guidance, refer to the BSH guidelines on transfusion for fetuses, neonates and older children (New et al., 2016). The Blood Safety and Quality Regulations (BSQR SI 2005 No.50 as amended) define blood components as a therapeutic constituent of blood [red blood cells, platelets, fresh-frozen plasma (FFP), cryoprecipitate or granulocytes], whereas blood products are derived from the whole blood or plasma [e.g. solvent detergent (SD)-FFP, albumin and anti-D immunoglobulin] and are classed as medicinal products. Blood components are excluded from the legal definition of medicinal products (The Human Medicines Regulations, 2012) and must be ‘authorised’ rather than ‘prescribed’ (Pirie & Green, 2008). Because SD-FFP is classed as a product, it must therefore be prescribed. However, the rest of the administration process for SD-FFP should follow the same guidelines as for blood components. It is imperative that patients are robustly and correctly identified at all stages of the transfusion process and that all spelling and number sequences on the patient ID band and all transfusion-associated documentation are correct and legible. For the purpose of this guideline, the term ‘patient ID band’ will be used to include all other terminologies (e.g. wristband) and any other forms of risk-assessed patient ID (e.g. photograph ID cards). Additional identifiers, such as address, middle name or gender, may be required according to the local policy and should be considered wherever core identifiers are referred to throughout this document. Initial patient ID on arrival to an outpatient or inpatient facility is a key step, ensuring that the right patient is issued with the correct ID band (Davies et al., 2006; Callum et al., 2011). There should be robust local policies governing the production and application of patient ID bands. It may be deemed difficult to use traditional ID ‘wristbands’ in certain patient groups (e.g. neonates or patients undergoing surgery where access to upper limbs may be restricted). It is strongly recommended that all patients receiving a transfusion are positively identified using an accessible patient ID band (or risk-assessed alternative), which is securely attached to the patient. If the patient ID band is removed for any reason, it should be the responsibility of the person who removed it to replace it without delay. Electronic systems are available to help improve patient ID procedures. The patient core identifiers may also appear on the patient ID band in an electronically readable format (e.g. bar code or radio frequency ID). Although information technology (IT)-based solutions generally offer improved patient safety (Davies et al., 2006; Askeland et al., 2008; Aulbach et al., 2010; Murphy et al., 2012; Nuttall et al., 2013), they should be robustly designed and implemented to ensure that patient safety is enhanced and that the systems are not open to any unintended risks and do not enable users to inappropriately override IT instructions (Davies et al., 2006; Patterson et al., 2006; Litchner et al., 2010; Andres et al., 2011; Murphy et al., 2012; Nuttall et al., 2013; McDonald, 2014). Further information on the use of IT in transfusion practice is available in the BSH guidelines for the specification, implementation and management of IT systems in hospital transfusion laboratories (Jones et al., 2014). In order to continually improve transfusion IT systems, all significant adverse events, errors or identified risks should be reported to SHOT and/or the appropriate regulatory authority. In such cases, as a minimum, at least one unique patient identifier, often a temporary emergency ID number, together with the patient's gender, must be used. There must be robust mechanisms for the issue and later withdrawal of these temporary ID numbers and for the subsequent linking of any duplicate patient's clinical and transfusion laboratory records. Further details are available in the BSH guidelines for the haematological management of major haemorrhage (Hunt et al., 2015) and the BSH guidelines for pre-transfusion compatibility procedures in blood transfusion laboratories (Milkins et al., 2013). When temporary emergency ID numbers are used, organisations should issue these as non-consecutive numbers [National Blood Transfusion Committee (NBTC), 2006]. Consecutive ID numbers invariably rely on a single-digit difference between often sequentially admitted and sometimes adjacent patients in the emergency department. It is imperative that the patient ID is checked and confirmed as correct at each stage of the transfusion process. All patients should, whenever possible, be asked to state their full name (first and last names) and date of birth. Where local policy requires, they should also be asked to state their address and/or spell their name. This must match exactly the information on the patient ID band and any other associated documentation required at that stage of the transfusion process. The patient's unique ID number on their ID band must also match exactly. This is the ‘positive patient ID check’. For paediatric transfusions, it is acceptable to ask the child to positively identify themselves if the child is deemed able to respond competently. For any patient who is considered unable to identify him or herself (e.g. paediatric patients who are unable to respond competently, unconscious or confused patients, or where there is a language barrier), additional care in patient ID must be taken and local hospital ID policies should be developed and followed. For example, some organisations may permit verification of the patient ID to be obtained from a parent or carer (if present at the patient's bedside). All organisations where blood transfusions take place should have a local policy or guideline that outlines every step in the transfusion process. Patient's clinical records should be available within the clinical area at the time of the transfusion. Minimum transfusion documentation in the patient's clinical records should include the following: Administration: All transfusion-related documentation should include the patient core identifiers. The use of specifically designed transfusion care pathways to record this information is encouraged. Errors in the transfusion process often involve inadequate communication (SHOT: Bolton-Maggs, 2015). Clear and unambiguous communication between individual staff, departments and different organisations, as well as with the patient, is a fundamental principle of safe and effective care. Wherever possible, communication should be in written format to minimise the risk of misinterpretation or transcription errors that may occur with verbal communications. There is a general consensus that regular training of all staff involved in the blood transfusion process (for the processes they are involved in) is vital for transfusion safety. There is very little evidence to indicate the optimum frequency of training in order to maintain knowledge and skills. We advocate a risk-assessed approach, tailoring training to staff groups taking into account factors, such as frequency of transfusion in their clinical area, individual practitioner's reflections on their own competency status (concerns should be discussed with their manager and/or other relevant authority) and transfusion-associated error rates. Therefore, the intervals between training may be different for individuals, various staff groups or clinical areas, but should be at least three yearly. Previous BSH guidelines recommended that all staff involved in the transfusion process should undertake three-yearly observed competency assessments. Evidence for this was low and based on recommendations made by the National Patient Safety Agency (NPSA, 2006). Since this time, SHOT has continued to analyse transfusion-related incidents and errors and has reported that despite individuals being competency assessed, transfusion errors continue. Transfusion committees across the UK have reviewed practices and have made their own recommendations for training and competency assessment. Their current recommendations are summarised in Table 1. In addition to this, the BSQR (SI 2005 No. 50 amended) requires that all staff involved in the collection and distribution of blood components undertake regular competency assessments. The Medicines and Healthcare Products Regulatory Authority (MHRA) who are responsible for monitoring compliance with these regulations in the UK have stated that the assessments for staff collecting blood components should be ‘undertaken in a manner that is commensurate with the level of risks associated with the process but should be at least every two years’ (MHRA Blood Consultative Committee, 2008). All staff involved in the transfusion process should receive training no less frequently than three yearly (two yearly for blood collection). Knowledge and understanding assessment should be performed at least every three years (two years for blood collection). Annual competence-based educational updates for staff involved in collection of blood from the issue fridge/local storage facility, with assessment of competence as agreed locally in compliance with the BSQR (2005). Biennial training for other staff involved in sampling and/or administration of blood. Continued monitoring of competence of all relevant staff through the appraisal process. NHS Quality Improvement Scotland (NHS QIS)* Clinical Standards for Blood Transfusion (2006) stated a two-yearly theoretical knowledge training requirement for all staff involved in the transfusion process. In 2015, NHS Healthcare Improvement Scotland (NHS HIS)* reviewed these improvement standards against current practice/guidance and archived the standards. It has been recognised that the two-yearly training/updating requirement for all staff involved in the transfusion process is embedded in NHS Scotland Health Boards supported and reported on by Better Blood Transfusion to the Scottish Transfusion Advisory Committee (SCTAC). *NHS QIS has been replaced by NHS HIS Minimum training requirements for clinical and support staff involved in transfusion practice should be three-yearly knowledge updates in ‘Right Patient Right Blood’ (Safe Blood Transfusion). Three-yearly training in blood components and their indications for use for doctors who authorise (prescribe) blood components. Healthcare staff should update their knowledge of transfusion that is relevant to their clinical practice and be able to demonstrate evidence of this in their appraisal documentation. Competency assessment of relevant staff, e.g. newly qualified staff. View evidence of competence for new employees assessed elsewhere. Practical competency assessment in the transfusion roles practiced by staff must be done at a minimum of three yearly intervals. Staff involved in the collection and delivery of blood components should undertake competency-based assessment for this role at least every two years. The decision to transfuse must be based on the clinical assessment of the patient and his or her individual needs. The reason and rationale for the decision to transfuse and the specific blood components to be transfused should be documented in the patient's clinical record. A more detailed guidance for the appropriate use of blood components is available in other BSH guidelines and in the Handbook of Transfusion Medicine (Norfolk, 2013). There have been a number of incidents reported to SHOT related to transfusion-associated circulatory overload (TACO) and overtransfusion, particularly in vulnerable patients (including patients with low body weight, patients more than 70 years of age and those with concomitant medical conditions that predispose to TACO, including cardiac failure, renal impairment, hypoalbuminaemia and fluid overload). The clinical assessment of the patient should include an evaluation of the patient's age, body weight, symptoms and concomitant medical conditions when determining the volume and rate of a transfusion and whether a diuretic should be prescribed. SHOT recommends that a formal pre-transfusion risk assessment for TACO should be performed whenever possible as TACO is the most commonly reported cause of transfusion-related death and major morbidity. The SHOT checklist includes assessment of cardiac and renal risks and assessment of fluid balance. Identification of risk factors may enable mitigation by the use of prophylactic diuretics or postponing the transfusion (Bolton-Maggs, 2017). For patients identified at risk of TACO, a written request should be made that during the administration of blood components, specific attention should be paid to monitoring the patient for signs of circulatory overload, including fluid balance. Single-unit red blood cell transfusions are recommended [National Institute for Health and Care Excellence (NICE), 2015] for adults (or equivalent volumes calculated based on body weight for children or adults with low body weight) who do not have active bleeding, with further clinical assessment to determine whether additional transfusion is required. The concept that one unit of red blood cells in additive solution increases Hb by 10 g L−1 should only be applied as an approximation for a 70–80 kg patient. For adults of low body weight, a calculation of 4 mL kg−1 of red blood cells in additive solution will produce an approximate rise of 10 g L−1 (Norfolk, 2013). Paediatric transfusions should be calculated in mL kg−1 according to the BSH guidelines on transfusion for fetuses, neonates and older children (New et al., 2016) and be authorised in mL. The provision of patient information to patients who may have or who have had a transfusion is also recommended by NICE (2015). This guideline does not intend to provide specific guidance related to the intricacies of consent or treatment refusal. The recommendations from the General Medical Council (GMC) for adults and for children 0–18 years of age should be adhered to (GMC, 2007; , 2008). The principles of consent for transfusion, as detailed by SaBTO (2011a), should be implemented for adults and children. Where transfusion of all or specific blood components is refused, or an Advance Decision (Living Will) exists, this should be documented in the patient's clinical records and communicated to all relevant HCPs. Local procedures detailing specific actions to be taken should be developed and followed. Blood components are excluded from the legal definition of medicinal products (The Human Medicines Regulations, 2012) and must be ‘authorised’ rather than ‘prescribed’ (Pirie & Green, 2008). As such, this guidance refers to authorisation of blood components and not prescription. There are no legal barriers to any appropriately trained, competent, locally designated and approved registered regulated HCP being able to authorise blood component administration. Ideally, to prevent communication or transcription errors, blood components should be authorised by the HCP making the decision to transfuse. Blood components should not be prescribed ‘pro re nata’ (PRN). Requests to the transfusion laboratory for the selection and release of blood components require a clear unambiguous and timely communication between the requester and the transfusion laboratory. Wherever possible, communication should be in written or electronic format to minimise the risk of misinterpretation or transcription errors that may be associated with verbal communication. Where requests for components are urgent, there should be a mechanism for alerting the transfusion laboratory, which includes real-time communication. Guidance for communication with the laboratory in the event of major haemorrhage is detailed in the BSH guidelines for the haematological management of major haemorrhage (Hunt et al., 2015). Known haemoglobinopathy, Transfusion laboratories should have a request form acceptance policy detailing actions to be taken if minimum requirements are not met. Incorrect or inadequate patient ID, leading to a sample for blood grouping being taken from, or labelled for, the wrong patient may result in ABO-incompatible transfusions, a third of which have resulted in serious complications or death (SHOT: Bolton-Maggs, 2014). All patients having a blood sample taken must be positively identified. The collection of the blood sample from the patient and the subsequent completion of details on the blood sample tube must be performed as one continuous, uninterrupted event at the patient's (bed)side involving one patient and one trained, competent and locally designated member of staff. The core patient identifiers on the sample tube must match those on the request form and the patient ID band. In the outpatient or community setting, patient ID bands may not always be used at the time of sampling. However, patients must still be positively identified using a risk-assessed alternative. Sample tubes must never be pre-written or pre-labelled. Pre-printed labels (pre-printed away from the patient or taken from the patient's notes, e.g. ‘addressograph’ labels) must not be used to label transfusion blood sample tubes for compatibility testing. Only labels that are printed ‘on demand’ next to the patient and immediately attached to the sample tube at the time of phlebotomy by the individual who took the blood sample are acceptable. Local policies or guidelines should detail how blood samples should be transported to and received by the transfusion laboratory in a timely manner, including emergency and urgent samples. Laboratories should also have a blood sample acceptance policy, which covers acceptable labelling of samples, as recommended in the BSH guidelines for pre-transfusion compatibility procedures in blood transfusion laboratories (Milkins et al., 2013). These guidelines also recommend that unless secure electronic patient ID systems are in place, a second confirmatory blood sample should be requested for confirmation of the ABO group of a first-time patient prior to transfusion, where this does not impede the delivery of urgent blood components. This is also recommended in the BSH guidelines on transfusion for fetuses, neonates and older children (New et al., 2016). The minimum sample tube information requirements are as follows: Removal of the incorrect blood component from storage location continues to be a major source of error in the transfusion process (SHOT: Bolton-Maggs, 2015). Many collection errors occur when the patient details on the laboratory-generated label attached to the blood component pack are not checked against the patient ID. This activity is subject to the requirements of the BSQR standards (SI 2005 No. 50 as amended), and only trained, competent and locally designated staff may collect blood components. There must be a clear audit trail of the collection, delivery and receipt (and return) of all blood components. Blood components must only be collected and received by a trained, competent and authorised member of staff. Because of the potential for transcription errors, telephone requests for the collection of blood components should only be permitted following a formal risk assessment. The staff member collecting the blood component from the storage location should use locally agreed specific documentation (e.g. blood component collection form or transfusion record), containing the patient's verified core identifiers and details of the component to be collected. These details must be checked against the details on the laboratory-generated label attached to the blood component pack. Also check whether it is the correct component type and the expiry date. If transport containers are used, they should be specifically designed and validated for blood components and must only be packaged for transport by trained, competent and locally designated staff. SHOT (Bolton-Maggs et al., 2013) notes the increased risk of a wrong transfusion when multiple units for more than one patient are collected. Unless rapid transfusion of large quantities of blood components is needed, or if components are being transported to remote areas using a transport container, only one unit should be collected for one patient at a time. The component should be delivered directly to the clinical area. Upon receipt in the clinical area, an appropriately trained and competent member of staff should check that the correct blood component has been delivered and record the date and time it arrived. In emergency situations, it may be necessary to provide emergency group O red blood cells that are not specifically prepared for the patient concerned. The issue of emergency stock must be controlled and documented so that patient safety and audit trails are not compromised. To assist traceability and ensure rapid replacement of emergency units so that future demands are not compromised, the transfusion laboratory must be informed immediately when emergency units are collected. Red blood cell units designated as emergency stock should be clearly identified as such and separated from matched and patient-labelled red blood cell units in the same refrigerator. Further details are available in the BSH guidelines for the haematological management of major haemorrhage (Hunt et al., 2015). All unused components should be returned to the transfusion laboratory according to the local policy. This is discussed further in Appendix 1. If less than 30 min, the unit may be returned directly to the issue location (e.g. in the designated storage refrigerator) according to the local policy. Although it is best practice that the number of occasions this occurs should be limited, there is no restriction on how often this may occur. Guidance from JPAC and BSH should be followed. JPAC (2013) states that once thawed, the component should be transfused as soon as possible. If delay is unavoidable, the component may be stored and should be used within 4 h if maintained at 20–24 °C, 24 h if stored at 2–6 °C or within 120 h stored at 2–6 °C only for patients who develop unexpected major bleeding (e.g. trauma) and for whom delay in providing FFP is detrimental (BSH: Green et al., 2016, JPAC, 2015, JPAC, 2016b), but it should be borne in mind that extended post-thaw storage will result in a decline in the content of labile coagulation factors. Based on the pragmatic expert opinion, JPAC (2016b) and BSH (Green et al., 2016) currently advise that thawed FFP that is out of a temperature-controlled environment (2–6 °C) can be accepted back into temperature-controlled storage (2–6 °C), if this occurs on one occasion only of less than 30 min. Transfusion of FFP should be completed within 4 h of issue out of a temperature-controlled environment. Any units returned after more than 30 min can be reissued for use only if the subsequent transfusion will be completed within this 4-h period. Guidance from JPAC should be followed. JPAC (2013) currently states that once thawed, the component should be transfused as soon as possible. If delay is unavoidable, the component may be stored and should be used within 4 h if maintained at 20–24 °C or 24 h if stored at 2–6 °C, but it should be borne in mind that extended post-thaw storage will result in a decline in the content of labile coagulation factors. No other detail is provided regarding returning out of temperature units to temperature-controlled storage. Pharmaceutical manufacturing company's product information should be followed. At present, Octapharma is the only supplier of SD-FFP in the UK. Octapharma (2012) currently states that after thawing, Octaplas® can be stored for up to 24 h at 2–8 °C, or 8 h at room temperature (20–25 °C), before use. No other detail is provided regarding returning out of temperature units to temperature-controlled storage. Guidance from JPAC should be followed. JPAC (2013) currently states that thawed cryoprecipitate cannot be refrigerated and therefore has to be used within a 4-h period following thawing. Any units returned could only be reissued for transfusion within this 4-h period. Guidance from JPAC should be followed JPAC (2013 and 2016c) currently states that platelets should be stored at 20–24 °C with constant gentle agitation. No interruption to agitation should last greater than 8 h, and the combined interruption over the unit lifespan should be no longer than 24 h. When platelets are returned from a clinical area, the duration of interruption to agitation and time out of temperature-controlled storage should be considered. Returned units can be reissued, retaining the same shelf-life, provided that any interruption to agitation or temperature control is no greater than the current JPAC guidance. These components, which are issued on a named patient basis only, are kept at room temperature (20–24 °C) without agitation (JPAC, 2013). If returned, units can be reissued (to that named patient only) within the lifespan of that unit. Failure to correctly undertake the formal identity check of the blood component with the patient prior to administration puts patients at risk of receiving the wrong blood (SHOT: Bolton-Maggs, 2015). Blood components must only be administered by trained, competent and locally designated registered regulated HCP. Consideration must be given to whether there are sufficient staff available to monitor the patient throughout the transfusion episode, by individuals who are trained in the management of transfused patients, including transfusion reactions and the emergency treatment of anaphylaxis (BSH Guideline on the investigation and management of acute transfusion reactions: Tinegate et al., 2012) and that resuscitation facilities are available. ‘Out of acute hospital’ settings should be risk-assessed regarding available facilities and resources, including emergency support. Transfusions should be given with the same attention to patient observations whatever the time of night or day. SHOT (Bolton-Maggs, 2015) has recommended that transfusions at night must proceed where there is a clear clinical indication and where there are sufficient staff to permit safe transfusion, including all required patient observations recommended in this guideline. However, it is important to recognise that there is a greater risk of cl