Transfusion errors and management

“A man’s errors are his portals of discovery” James Joyce Error is an inherent component and an intrinsic by-product of every activity in which humans are involved. This tendency to err is philosophically perhaps one of the defining traits that characterize us as human beings. Transfusion errors probably existed once blood transfusion became an essential component of medical treatment. However, the systematic acknowledgement, recognition and analysis of these errors are certainly a much more recent phenomenon. According to an Institute of Medicine (IOM) report released in 1999 [1], medical errors are one of the leading cause of death in the US and have caused an estimated 44 000–98 000 deaths per year, which is soberingly more than the motor vehicle accident death rate of 43 458 for that particular year. Errors and adverse events carry a high financial burden. The Agency of Healthcare Research and Quality (AHRQ) estimated that the cost of medical errors resulting in adverse events is approximately $37·6 billion per year, and $17 billion of those costs are preventable [1]. Significantly, the AHRQ has reported that medical errors are not only because of medication errors but can result from blood transfusion errors, misdiagnosis, misinterpretation of laboratory results with incorrect therapy, equipment failure, hospital acquired infections or misinterpretation medical orders. Looking specifically at transfusion mortality, Schmidt reviewed 69 records of deaths labelled as transfusion mortality and concluded that 31 (44·9%) were directly attributable to transfusion, of which 25 were due to red cell incompatibilities and discrepancies [2]. Myhre reviewed 113 records fatalities reported to the Food and Drug Administration (FDA) as a sequelae of blood or blood product transfusion from 3 April 1976 to 31 December 1979 [3]. Thirty-three fatalities were due to post-transfusion hepatitis, and three involved plasmapheresis or leucopheresis donors. The remaining 77 cases were of three categories: the vast majority were preventable errors: 47 cases (61%) were due to “clerical errors” such as drawing the wrong blood specimen or giving the wrong unit of blood to the patient; 8 cases consisted of genuine errors that occurred in the laboratory and 22 miscellaneous cases. Many other recent reports exist with the majority retrospective and involving registry data (e.g. reporting to the FDA). Honig and Bove listed 70 fatalities, 38 of which were due to transfusion errors [4]. Camp and Monaghan studied 126 FDA reports from 1976 through 1980. They found that 64 reactions that were due to simple clerical errors, such as giving the blood to the wrong patient [5]. In a landmark paper, Sazama and colleagues studied transfusion-related deaths reported to the US FDA for 10 years and identified 131 fatal ABO incompatible transfusions. The study showed that the most frequent error leading to a fatal outcome was administration to someone other than the intended recipient [6]. Records of transfusion errors in the New York State for the past 10 years (Linden) showed that erroneous blood administration was observed in 1 out of 19 000 RBC units administered. Half of these events occurred outside the blood bank (administration to the wrong recipient, phlebotomy errors, testing of the wrong specimen, transcription errors and issuance of the wrong unit) with 15% of cases involving multiple errors [7]. McClelland and Phillips conducted a study and analysed the data from haematology laboratories in the UK for the years 1990 and 1991. They sent a questionnaire to 400 laboratories involved in blood transfusion and only 245 (61%) submitted their reply. The reports included data on deaths, morbidity and near misses. One-third of the laboratories reported incidents in which the patient received a wrong unit of blood [8]. Errors can be defined in a number of ways, but most would agree that the crucial and distinguishing characteristic involves a failure (deviation) in the performance of a standard operating procedure (SOP) resulting in an unintended and unwanted consequence. Many transfusion errors can be benign but may be pointers to a careful re-assessment of the process. Others, however, can result in catastrophic consequences for the patient; particularly if it involves transfusion of ABO incompatible blood. Errors can happen at any point along the transfusion chain (vein to vein) but particularly in the following points, which are considered the weakest links in the chain [9]: Decision to transfuse Sample errors Laboratory errors Blood issue and administration errors A decision to transfuse should be based upon relevant clinical signs and symptoms supported by a laboratory result. Although the decision to transfuse is ultimately a clinical one arrived at the bedside, there is now a vast body of concordant literature on guidelines and thresholds for transfusion of each of the relevant blood components. Inappropriate, unnecessary and nonevidence-based decisions to transfuse may result in more harm than benefit to a patient. Although traditionally not looked upon as a “classical” transfusion error, this inappropriate usage of blood may actually be the largest source of preventable error that would benefit from further systematic scrutiny. A patient succumbing to TRALI from an unnecessary transfusion of a unit of fresh-frozen plasma (FFP) should be considered a possible preventable transfusion error. Education of the hospital physician is key and a close liaison between the transfusion medicine specialist and the hospital will go a long way to achieving this goal. The establishment of hospital, local or national guidelines on transfusion indications should also be viewed as a priority in all healthcare services. A strong, enthusiastic and active hospital transfusion committee is also important in this respect. In deciding to transfuse, failure to provide the transfusion laboratory with details regarding the patient’s transfusion history or special blood requirements can result into serious morbidity and even mortality. These important details include a previously detected alloantibody, requirements for washed units, leucoreduced or irradiated blood components. Conversely, hospital transfusion laboratories should also have a system for recording special needs for patients including significant alloantibodies and the need to irradiation This should as far as possible; be robust and independent of the need for the requesting physician to continually indicate on the transfusion order form. The next critical stage of the transfusion process is that of blood sampling for pretransfusion testing. Sample errors may be due to wrong labelling of sample tubes or collection from the wrong patient (wrong blood in tube). Unsafe practices include labelling tubes away from the bedside, failing to check patient identity or the use of pre-labelled containers. Hospital SOPs should be in place and stringently enforced with regular audits to check on performance and compliance. Poor techniques in blood sampling for diagnostic investigations may also give rise to inappropriate transfusion, sometimes with clinically significant consequences. The UK Serious Hazards of Transfusion (SHOT) reported cases wherein blood were taken from a “drip arm,” resulting in an erroneous haemoglobin result and an inappropriate decision to transfuse contributing to the deaths of two patients [10]. Approximately 30% of “wrong blood” events reported to SHOT arise in the blood bank laboratory, which is equivalent to what Linden et al. reported [7]. A disproportionately high number of laboratory errors took place outside of “core hours” and at night, when staff are fewer in number, may be relatively inexperienced and working under pressure. Unless the transfusion laboratory is appropriately staffed throughout the 24-h period, requests for transfusion at night should be restricted to urgent and emergency cases [9]. Errors at these stages constituted 40% of “wrong blood” events reported to SHOT in 2003 and resulted in 12 ABO incompatible transfusions. Anecdotal case reports provide insights into the system failures. Inaccurate verbal instructions and the common pitfall of similar patient names can contribute to blood issue and administration errors. Transport or transit errors can occur during the transfer of the blood and blood components from wards to operation theatres or vice versa, including not storing at adequate temperatures [9]. Because of the certainty of human errors, it is important that there is a system in place to identify all errors occurring at any stage of the transfusion chain. A near miss is any error that, if undetected, could have resulted in harm to the patient as a consequence blood product, but was recognized before transfusion occurred. Near misses reflect more closely the true incidence of transfusion errors and are valuable in realigning efforts and cost with risks. Errors should be analysed, grouped and characterized in a consistent fashion, so logic-based and systematic solutions can be implemented. This is much preferred over a series of reactive, episodic and unplanned temporary quick fix approaches to remedy the problem. There is disproportionately more published data regarding adverse event reporting because of errors when compared with near miss reporting. In reality, we could gain more from near miss reporting because they occur more frequently and useful data could be collected for analysis to prevent future adverse event occurrence. This is the crucial distinguishing feature of a near-miss event: the “recovery step” preventing future harm. An example of an event-reporting system is the medical event reporting system for transfusion medicine (MERS-TM), which is currently being used in more than 30 sites internationally. The MERS-TM was developed in 1995 through a grant by the Heart, Lung and Blood Institute in the US. The system ensures complete reporting of errors and consistent application of rules of analysis of those data. The MERS-TM is divided into different components: detection, selection, description, classification, computation and interpretation. The event is then classified by using preset event codes and 20 possible causal codes. Causal codes are subdivided into latent failures (organizational and technical), active failures (human-related errors) and patient-related factors [11]. A systematic capture and analysis of the entire transfusion process in terms of its risk benefit, error occurrence, adverse events and sequelae has rightly captured the attention of the transfusion community in the last 10 years. This concept of haemovigilance is defined by the International Haemovigilance Network as “a set of surveillance procedures covering the whole transfusion chain intended to collect and assess information on unexpected or undesirable effects resulting from the therapeutic use of labile blood products, and to prevent their occurrence or recurrence” [12]. Two of the earliest national surveillance schemes to be set up, those of the UK and France, clearly demonstrate fundamentally different approaches to haemovigilance and have been the benchmarks by which other countries have based their own schemes. Haemovigilance was established in France as a national system of mandatory surveillance which encompasses the entire transfusion chain. Considerable resources are spent on this with a sizeable workforce dedicated to this process. As such, a wealth of data over the years has emerged [13]. Serious Hazards of Transfusion is the UK’s haemovigilance system [10]. Compared with the French system, SHOT is voluntary and only captures reports on serious cases and hence the incidence of risks from both systems may not be directly comparable. From 1996 to 2003, there were 226 cases of ABO incompatible transfusions that were reported to the SHOT scheme. During this period, 23 million blood components were issued by the UK blood services and 2087 adverse events analysed, of which 1393 (67%) were reports of “incorrect blood component transfused” (IBCT) where a patient received a blood component intended for another patient or that did not meet appropriate specifications. The risk of an error occurring during transfusion of a blood component is estimated at 1:16 500, an ABO incompatible transfusion at 1:100 000 and the risk of death as a result of an IBCT is around 1:1 500 000 (SHOT, 2003). These figures may be underestimates, as not all adverse events may be recognized and not every hospital reports to SHOT [9]. Many more countries have either implemented or are in the process of implementing haemovigilance systems based on one of the above models. Collaborative groups now exist which greatly facilitates the exchange of information, standardization of definitions and harmonization of practices. One such grouping is the European Haemovigilance Network (now known as International Haemovigilance Network) which include countries within Europe as well as countries outside Europe (Singapore, Canada, Australia, Japan and the US) [14]. There were 24 cases of incorrect blood transfused reported to the Singapore National Haemovigilance Program since its conception in 2002 (Figure A). This is probably not the true incidence and may reflect under reporting of cases from hospitals as the system is voluntary. More than half of the cases (54%) involve ABO compatible transfusions (e.g. patient receiving the wrong that is intended for another patient but is ABO compatible) and special requirements not met (e.g. failure to irradiate blood products). In most of these cases, patients did not experience any untoward incident after the transfusion. Nonetheless, the same error could have easily resulted in fatality. There were four cases of ABO incompatible transfusions reported and two resulted in significant clinical harm to the patient. The first case involved transfusion of several units of group O FFP to a group A patient which resulted in a haemolytic transfusion reaction. Analysis of the events leading to the incident showed that the error happened in the laboratory during release of the FFP units. The second case was an ABO incompatible red cell transfusion (A-positive blood to an O-positive patient). The patient was successfully resuscitated and both events reported to the Ministry of Health as sentinel events. The other two cases were small volume (<500 ml) transfusion of ABO incompatible FFP and patients remained clinically well with biochemical markers of some haemolysis [15]. There was a steady increase in the number of near miss reported from 2002 to 2007 (Figure B). Near-miss events represent 18% of the total number of cases reported to the Singapore National Haemovigilance Program with an incidence of 3·2 errors per 100 000 units when compared with 7·0 errors reported by Jørgensen and Taaning in the Danish Registration Transfusion Risk from 1999 to 2002 and 5·1 in the UK SHOT report from 1997 to 2001 [16]. This increase in reports may be due to increased awareness and implementation of audit measures resulting in previously undetected errors being identified and now reported. More than half the cases of near-miss incidents reported were sample errors (Figure C), which is similar to that reported in other haemovigilance programmes. Local data showed that majority of sample errors occurred because of failure to adhere to hospital protocols for patient phlebotomy. As a result, the Singapore National Haemovigilance Program emphasized in its annual report in 2007 that all personnel involved in the processing, collection and administration of blood should strictly follow SOP and undergo appropriate training particularly in patient identification. Data collected in 2008 did show an improvement and a decrease in cases reported although the causal relationship is difficult to establish. Solutions that could reduce errors thus enhancing transfusion safety are available and can often be adapted from the airline and automotive industries where it has been in use for years and where the tolerance for error margins are particularly slim. The pioneering airline industry, for example, promotes a culture of safety and openness in talking about errors. It also invests resources in process (flight) as well as component (aircraft) safety; and uses direct observational in-flight audits, a non-punitive industry-wide error-reporting system, and technology such as in-flight recorders to analyse near-miss events [17]. Another useful partner is the long established pharmaco-vigilance system. Medicine errors have long been recognized and have been systematically captured. The process of error reporting resulting in process change, physician education and greater knowledge of the drug in question are key areas which haemovigilance can learn from. In every institution that is involved in the collecting, testing, processing and administration of blood, a quality program must be in place for the identification and the management of errors. A quality culture must exist that emphasizes the responsibility of everyone involved as well as a non-punitive approach that encourages frank reporting and open discussions without any fear of retribution or punishment. Staff should be made aware of their responsibility to ensure the quality of blood at all times and the quality programme must provide guidance on policies, directions, documentation and other related issues [18]. Clearly written and unambiguous procedures and protocols should be readily accessible at the workplace to the staff at all times to prevent errors. These SOP and protocols provide step-by-step directions on how to perform job tasks and procedures. Safe transfusion therapy is also dependent on a skilled, professional laboratory and medical workforce. Training must be provided for each procedure for which employees have responsibility. Medical personnel, including physicians, nurses and students, should be provided with adequate information about transfusion and blood safety. To ensure that skills are maintained, regularly scheduled competence evaluation of all staff whose activities affect the quality of blood, components, issues or services should be conducted. On a basic level, staffing numbers are also important so that the service is not stretched to a “fractured” level. An improved standardized system for labelling blood and blood components is also important to minimize human error in transfusion medicine. An example of this labelling system is the ISBT 128. It is an alphanumeric, high-density, widely supported symbology that can and reduce errors. that have not implemented to ISBT may not have the to blood with ISBT or will have to to which are more error and technology that is in or have made its way to the hospital and are now being used to prevent transfusion errors. The SHOT has the evaluation of transfusion and technology for that the unit of blood is administered. The system involves the use of a and that essential first of and hospital number technology can where the use of technology may be technology is being used in and may or information the of blood and blood information such as clinically significant red cell or adverse reactions to or blood transfusions can be if are in patient’s It is also useful in the operating where patient identification are difficult without technology can all these information patient information on a similar to a Compared with the the technology for use by the transfusion therapy is a process that the and the different hospital including the clerical services and services. The Hospital Transfusion has a key in this culture of safety throughout the hospital and all infections have a in blood the and management of errors is often not emphasized contributing to patient mortality and The consequences of transfusion errors are and this has to out of blood and into the hospital where the transfusion should be into at systems and Near miss reporting and Haemovigilance schemes have clearly identified priority areas in need for to transfusion safety.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.