NHO Report 2008 2009 - Irish Blood Transfusion Service

McSweeney, (IBTS) Dr. Patrick Hayden (GalwayUniversity Hospital) Ms. Gretta Boyle (HVOConnolly Hospital) focused on developingaspects of haemovigilance and the NHO team,Dr. Lawlor, Marina Cronin, Roisin Brady and JohnCrumlish (IBTS) presented a breakdown of thehaemovigilance reports submitted for 2008.Dr. Emer Lawlor, NHO Director when presentingthe prize, commended the high standard ofposters displayed. The winning entry ‘Apheresisversus Pooled Platelets’ was submitted by Ms.Anne Thompson and the Haemovigilance Team inOur Lady’s Children’s Hospital Crumlin which setout details of an audit undertaken to identify andcompare the number of reactions caused byPooled and Apheresis platelets over a four and ahalf year period. The results showed a decreasedreaction rate to pooled platelets since the IBTSintroduced Platelet Additive Solution (PAS) in July2007.The event attracted in excess of 190 delegatesdrawn from medical, nursing and scientificbackgrounds throughout Ireland and abroad.From the evaluations, comments and feedbackreceived, all attending enjoyed the event.Both years there was a general consensus thatthis was an excellent opportunity to meet likemindedcolleagues and develop network contactswith others working in the area ofhaemovigilance. Some of the suggestionsreceived will help with the design of theprogramme for future conferences and IBTSHospital Liaison Days.The NHO team wishes to thank all involved inmaking arrangements for the conferences and fortheir assistance during the event. We especiallythank those working within the Haemovigilancenetwork who promoted the event with theircolleagues. Those who contributed their time andeffort in chairing sessions and who presented areespecially acknowledged. The support and cooperationof the management and staff at our twovenues, the Castletroy Park Hotel, Limerick andthe Royal Hospital Kilmainham, Dublin is alsoacknowledged.Presentations at International meetingsEHS 2008The European Haemovigilance Seminar (EHS) washeld in Frankfurt in 2008 and Dr. Lawlor wasinvited to present at this event on OverTransfusion and a poster was submitted to EHSentitled ‘An exploration of reported TransfusionAssociated Circulatory Overload (TACO) andTransfusion Related Acute Lung Injury (TRALI) inIreland 2000-2006’ presenting an analysis ofTACO and TRALI cases received by the NHO overa six year period. This confirmed that thereported incidence of TACO is up to 20 timesmore common than TRALI. The poster alsohighlighted the series of measures taken by theIBTS to reduce the risk of TRALI.In 2008, M. Cronin presented a paper onevidence based practice at the British BloodTransfusion Service Apheresis and Bloodcollection Special Interest Group in Manchester.In May 2008, the Society of Blood Transfusion inSpain invited the NHO to present at the NationalCongress on hospital based haemovigilance. M.Cronin presented a paper in Cadiz in Spain.EHS 2009 was held in Rome and Ms. JackieSweeney was invited to give an oral presentationon her abstract the ‘Annual Notification of SeriousAdverse Reactions and Events (ANSARE) Ireland2006-2007’ co-authored by Ms. Roisin Brady, Ms.Marina Cronin, Ms. Marcia Kirwan and Dr. EmerLawlor. The key finding from this research showedthat reports collected on ANSARE accounted for285 (54%) of the total 525 (46%) reports analysedby the NHO for this reporting period, for just overhalf of the total number of reports received, andunderestimate the overall rate of reporting asthey do not cover clinical errors.ASH 2009A poster entitled ‘Incidents and Relevant Aspectsof Transfusion Associated Circulatory Overload’co-authored by Dr. Andrea Piccin, Ms. MarinaCronin, Mr. Ciaran Murphy, Ms. Elva Eakins andDr. Emer Lawlor was displayed at the annualAmerican Society of Haematology (ASH) meetingheld in New Orleans in December 2009.The SHOT UK meeting was held in London in2009, and Ms. Marina Cronin, Ms. Kathleen Heeryand Dr. Emer Lawlor represented the NHO.A poster entitled ‘Education opportunities forblood donation and transfusion practitioners’presented a summary of the educationaldevelopments achieved by the NHO in DCU.6

Open Days and IBTS crossmatch ServiceAll newly appointed HVO are invited to the NHOOpen Day where the workings of the NHO areexplained, with particular emphasis placed onreactions and event reporting. One was arrangedin 2008 at which 22 people attended and anotherwas held in 2009 at which 14 people attended.Nationwide networking among HVO is alsopromoted through regular telephone/emailcommunication and personal visits.During 2008, new arrangements were put in placeto facilitate the electronic completion of theANSARE forms. HVOs were provided with anopportunity to attend an information and trainingsession, arranged by the NHO and facilitated bythe IBTS Information Technology (IT) Department.The purpose of this session was to familiariseHVOs with these arrangements and encourageparticipation with the new system. This system wascontinued in 2009, with all but five hospitalsavailing of this facility.E Learning ProgrammeThe E Learning programme in blood transfusionpractice was developed by the Effective Use ofBlood (EUB) Group of the Scottish National BloodTransfusion Service (SNBTS). The programmeconsists of three levels, with an additional modulefor those working in the blood transfusionlaboratory. It is aimed at practitioners working intransfusion practice and permits those unable toattend formal training sessions, to take part incontinuing education in blood transfusion practiceas well as enhancing face-to-face educationalsessions.In 2008 the programme platform moved from theORAS TM Gold to the LearnproNHS site hosted byLearnpro Ltd. Access to the site is funded by theIBTS. The NHO is part of the editorial group of thecontinuing education programme of the BetterBlood Transfusion for E Learning which reviews theprogramme content.Following the successful pilot project, the resultsand strategy for national roll-out were presentedat a one day event in June 2008 at the NBC.National implementation of the programmebegan in autumn 2008 and continued through2009. During roll-out, key stakeholders wereidentified at each hospital as well as a programmeadministrator (in most cases the hospital HVO) toimplement the project. Each administrator wasprovided with training material and invited toattend a one-day training session to assist them inrolling out the programme.In the latter part of 2008, three E-Learning trainingdays were held as part of the programmeexpansion. Seventeen representatives from sixteenhospitals attended for training. The programmewas extended to more hospitals during 2009.Working PartiesThe Better Blood Transfusion Network (BBTN) is aworking group of UK and Irish haematologists andtransfusion medicine specialists, hospital cliniciansand transfusion nurse specialists, set up to shareinformation on best practice in the clinical aspectsof blood transfusion. Dr. Emer Lawlor, Ms. MarciaKirwan and Ms. Marina Cronin represented theNHO and IBTS during 2008 with Ms. MarinaCronin also attending meetings held in 2009.There were three meetings held during 2008 inEdinburgh, Dublin and Cardiff and a further threeheld during 2009 in Bristol, Edinburgh and inDublin during the 10th Anniversary NHOcelebrations.Ms. Jackie Sweeney represented Dr. Lawlor at anEU Working Group Meeting on Serious AdverseEvents and Reactions, co-ordinated under theCommission of the EC. Ms. Donna Harkin of theIrish Medicines Board also attended this meeting.This meeting was convened to discuss the firstversion of the common approach document whichaimed to reach a consensus on the annual reportof Serious Adverse Events and Reactions.NHO NewsThe information newsletter NHO News iscirculated to HVOs to provide an informal forumto report initiatives from the NHO and individualhospitals, including local education and trainingevents that may be of interest to other HVOs.Details and events of national and internationalinterest are also reported. During 2008, threeeditions of the newsletter were issued and during2009, one edition was issued.Information on haemovigilance can be directlyassessed on the IBTS website at www.giveblood.ie(Clinical Services–Haemovigilance).7

IBCT /SAE Key Points andRecommendations for 2008 - 2009General Recommendations• Reporting of both mandatory and nonmandatory events make importantcontributions to patient safety and attentionneeds to be refocused on the clinical areas ofblood transfusion not covered by the EUDirective.- The reduction in reports from clinical areasprobably reflects the emphasis on therequirements of laboratory accreditationand mandatory SAE reporting. However,the slight increase in reporting from theclinical area in 2009 is encouraging.• The role of the bed side check as the lastbarrier to prevent transfusion errors must behighlighted to clinical staff. Several incidentsreported could have been detected during thepre-transfusion check. Failure to do so raisesconcerns as to the attention paid to this criticalstep, and highlights the importance ofcontinuous education programmes for clinicalstaff involved in blood transfusion practice.Training and education• Ensuring training of staff especially medicalstaff is difficult. Implementation of astandardised transfusion programme forundergraduate medical students should ensureemerging clinicians will have an understandingof safe transfusion practice.• Ongoing hospital based education deliveredby HVOs, Consultant Haematologist and seniormedical scientists is also critically important tointegrate theory and safe practice. Initiativessuch as audit, provision of feedback,presentations or acting as a “clinical/laboratory” presence by the ConsultantHaematologist, hospital HVO and seniormedical scientists in the hospital blood bank(HBB) are important in raising the profile ofsafe transfusion practice.• Hospital Blood Banks should ensure that asrequired by ISO 15189 there is adequatetraining and competency assessment of staff,particularly staff who do not routinely work intransfusion.Learning from Errors• Adverse event review and reporting is a verypowerful way of organised learning inorganisations in general and also in transfusionservices. The information gained from theidentification and analysis of adverse eventswill enable the identification of gaps in thetransfusion service and perhaps other servicesin the hospital which require attention. Thisdata can be used to identify trends andpatterns of events which reoccur and havepotential to cause harm to patients, andfacilitate development of appropriatestrategies to enhance patient safety(Commission on Patient Safety and QualityAssurance, 2008).• Use of a formal root cause analysis protocol willensure a systematic, comprehensive andefficient investigation, and will outrule thepotential of simplistic explanations and routineassignment of blame.• The NHO has worked with the clinical riskadvisors in the Clinical Indemnity Scheme (CIS)to ensure all haemovigilance staff receivessystem analysis/RCA training. Furthermore theimplementation of the recommendations of thepatient safety commission will include anational roll out of an agreed approach tosystems analysis to all health careorganisations.Changes to practice/Follow up action• Introduction of a change should includedevelopment of policies to support practicechange, informing all relevant stake holdersand provision of training to ensure that theinformation on change is disseminated andacted on.• Haemovigilance and transfusion servicesshould monitor these changes not only toevaluate the impact of the change in terms oftransfusion service, but the potential to impacton other hospital services. This follow-upmonitoring is crucial to ensure ongoinglearning and improvement and is characteristicof a quality service.8

Unnecessary transfusions• Underlying anaemia has been recognised as acause of unnecessary transfusion andincreased morbidity in patients undergoingelective surgery. A recent publication by theNetwork of Advancement of TransfusionAlternatives (Goodnough et al, 2010) maderecommendations on detection, evaluationand management of pre-operative anaemia.• Where several units are prescribed fortransfusion, patients Hb should be checkedbetween units. This will minimise risk ofunnecessary/over transfusion.Unnecessary Transfusion in nutritional anaemia• Each year, the NHO receives a number ofreports of unnecessary red cell transfusion inpatients with iron deficiency anaemia, and thisis likely to represent significant underreporting.• Asymptomatic patients with iron deficiencyanaemia should be treated with iron therapy.Oral iron should be continued for at leastthree months after deficiency has beencorrected so that iron stores are replenished.Ascorbic acid may enhance iron absorption.• Intravenous iron preparations should beconsidered in cases where patients have eitherpoor tolerance of oral preparations or thereare compliance issues. It normalizeshaemoglobin faster and more reliably thanoral iron.• Patients with megaloblastic anaemia respondvery rapidly to vitamin B12 / folate and veryrarely require transfusion.Unnecessary transfusion due to failure ofknowledge or lack of communication• Platelets should be given within one to twohours prior to a procedure, allowing formeasurement of post transfusion values.• Clinical teams caring for patients shouldcommunicate with the HBB to ensure patientsdo not receive blood componentsunnecessarily where procedures have to becancelled or postponed.Unnecessary Transfusion due to incorrectresults• Laboratories should ensure that validatedresults are available to clinical areas in a timelymanner to minimise the potential forunnecessary transfusion due to a delay inposting current haematology results or whereposted results were not validated on thelaboratory computer system (LIS).Near Patient testing• Near patient testing may be necessary inemergency settings. Where this is used,maintenance and validation of equipment aswell as ongoing training and competency ofclinical staff must be ensured. Hb resultsleading to transfusion should be checked inthe laboratory at a later stage.Selection of ComponentsCaution crossing ABO groups in plasma richcomponents• Although Group O donors are considered‘universal’ donors of red cells, Group Oplatelets have anti-A and anti-B in thesuspending plasma which can causehaemolysis in A, B or AB patients even if testsfor high titre haemolysins are negative.• Group O apheresis platelets should only beused for Group O patients. Group O pooledplatelets which are suspended in plateletadditive solution are less likely to beassociated with haemolysis if they have to beused for Group A or B patients where thepatient’s correct group cannot be provided asthey have reduced amounts of plasma.• Patients whose blood group is AB can betransfused with A or B red cells, but plasmacomponents which contain anti A or B shouldbe avoided.Transfusion of antigen incompatible red cells• Patients who have had previous pregnanciesor transfusions are at risk of developingantibodies. There should be robust systems todetect and reconcile patients’ previous9

histories and transfusion records. Very oftenthe HBB can be unaware of patients’ historyand the potential for antigen incompatibletransfusions can be high. Hospitals shouldhave policies covering the transmission of apatient’s antibody/transfusion history andspecial requirements when a patient istransferred to another hospital.• Where reports of investigations are receivedfrom the reference centre indicating thepresence of antibodies, HBB staff should checkthese against the patient’s current transfusionneeds, or in the case of an infant, the mother’santibody history, to ensure antigen negativeblood is provided.• Development of a national register of patientswith antibodies would reduce the risks oftransfusing antigen incompatible cells to thesepatients.The UK Transfusion Collaborative (Chaffe et al,2009) has recommended that the staffinglevels and skill mix should be adequate toensure the safe and effective delivery ofroutine and emergency services during allwork periods.Transfusion of incorrectly stored units• Best practice guidelines indicate that apatient’s status should be checked prior tobringing a unit of blood to the bed-side.Adherence to these guidelines will minimisethe risk of transfusing units which have beenout of controlled storage for too long andpotential wastage of a scarce resource.Paediatric Practice• Almost 21% of all reports of IBCT/SAE in 2008and 2009 related to paediatric patients.- Analysis of IBCT/SAE for both years interms of potential to cause harm topatients showed that a majority ofreported IBCT/SAE had high potential tocause harm in paediatric patients whencompared with reports in the adultpopulation, thereby highlighting risk topaediatric patients receiving transfusions.• Where care is shared between hospitals, thereshould be policies in place indicating theprocedure to follow for patients with specialtransfusion requirements.• Paediatric patients are long term survivors oftransfusion therapy. It is important that bothclinical and laboratory practitioners working inpaediatric centres continuously seek tominimise donor exposure for these patients.• Several of the paediatric SAEs were associatedwith on-call scientists with a busy workload.10

SAR Key Points and Recommendationsfor 2008- 2009Acute Non Hemolytic Transfusion Reactions(FNHTR ,AA)• Whenever possible, as a minimum, bloodcultures and investigations for haemolysisshould be taken on patients suffering a FebrileNon Haemolytic Transfusion Reaction (FNHTR)to exclude red cell incompatibility or bacterialcontamination.• Reaction Alerts in patient charts and/or on thehospital patient admittance system and ITsystem can be valuable in those patients witha previous Anaphylaxis/hypersensitivity (AA) orFNHTR reaction to ensure appropriatecomponent selection and pre medication priorto future transfusions.AHTR ( Acute Haemolytic TransfusionReactions)• In an emergency it may be necessary to issueleast incompatible blood for a patient butsamples should also be sent to a referencelaboratory for investigation and identificationof the antibody and to ensure a supply ofsuitable compatible units for ongoingtransfusionDelayed Haemolytic Transfusion Reactions(DHTR)• It is likely that Delayed Haemolytic TransfusionReaction (DHTR) is underdiagnosed. It isessential that any patient presenting withunexplained anaemia some days after atransfusion should be investigated forimmunological haemolysis (bilirubin, Lacticdehydrongenase (LDH), Direct AntiglobulinTest (DAT) and antibody screen) to excludeDHTR. In a number of the reports of DHTR in2008 and 2009 the investigation wasincomplete. The successful diagnosis alsodepends on accurate history taking and theeliciting of a history of recent transfusion.• There should be robust systems /policies todetect and reconcile patients’ previoushistories and transfusion records. If a patient istransferred to another hospital, theirantibody/transfusion history should betransmitted to the receiving hospital. This issupported by the recently published draftNational Standards for safer better health caredocument (HIQA 2010 ) which states thatservice providers share necessary informationto facilitate the transfer or sharing of care in atimely and appropriate manner• Patients who have had previous pregnanciesor transfusions are at risk of developingantibodies. Very often the HBB can beunaware of patients’ history and the potentialfor antigen incompatible transfusions can behigh. Development of a national antibodyregister could address this risk by ensuringaccess to patients’ antibody history. It wouldalso reduce the requirement for repeatlaboratory testing.- This would only be feasible with theimplementation of a national UHI, arecommendation made by the HIQA(2009) and supported by the NHO. A UHIwould facilitate improved and safer accessto patients’ records on a national antibodyregister thereby ensuring safer transfusionpractice for patients.Transfusion Associated Circulatory Overload(TACO)• Particular attention should be paid to patientswith underlying conditions which may increasetheir susceptibility to TACO. These include;- Elderly patients- Infants and children- Patients of low body weight- Patients physiologically compromisedparticularly those with a history of cardiacrespiratory or renal insufficiency or chronicanaemia.• Transfusion should be on a unit by unit basis,with a medical assessment of the patient priorto commencing transfusion and beforeadministering any further component. Thisassessment should include a;- careful estimation of the patient’shydration and cardiac status prior to thetransfusion,- thorough review of the patient’s fluidbalance during the transfusion- possible need for diuretic therapy as thiscan reduce the risk of TACO and may be11

Table 4: Breakdown of NHO incidents accepted 2000-2009 (n = 2127)Year IBCT AA TACO DHTR STTI TRALI PAD Unusual/ AHOSTR 1 TAD 2 Hypotensive Total/SAE Unclassified Reaction 22000 31 22 8 2 7 - - 1 14 852001 69 35 16 1 2 3 3 3 12 1442002 87 31 10 9 3 2 5 - 8 1552003 115 23 14 9 4 1 6 - 8 1802004 126 35 15 4 3 - 7 - 24 2142005 173 22 25 5 6 - 3 - 32 2662006 187(32 3 ) 29 34 4 8 2 0 0 40 3042007 115(32 3 ) 40 18 6 4 0 0 5 34 2222008 147(53 3 ) 41 39 4 8 1 0 6 40 2 2 2902009 157(46 3 ) 28 18 14 4 0 0 3 39 3 1 267Total 1207 306 197 58 49 9 24 18 251 5 3 21271Prior to 2006 Acute Transfusion Reactions (both Acute Haemolytic Transfusion Reactions and Febrile Non Haemolytic Transfusion Reactions) werereported as Acute Haemolytic or Other Severe Transfusion Reactions (AHOSTR).2Collection of these reports commenced only in 20083Denotes Mandatory SAE. Two blood establishment errors were also included on ANSARE.The incidence of SAE /IBCT and SAR per unitdistributed from the IBTS in 2008 and 2009 can befound in Appendix 1.Annual Notification of Serious Reactions andEvents ( ANSARE)Commission Directive 2005/61/EC Annex II D andIII C require reporting establishments to completethe Annual Notification of Serious AdverseReactions and Events (ANSARE) form whichcollects mandatory SAE and SAR.The majority of reporting establishments submitthe report electronically with only smaller facilitiesopting to report by hard copy. The number offorms received varies slightly each year. In someinstances, where a reporting establishment acts asa blood bank for another site, only one ANSAREform is returned. Other reporting establishmentschose to submit a separate ANSARE, despitereceiving blood components from another site.Finally some hospitals may no longer transfusepatients.14

ANSARE 2008Seventy-six ANSARE forms were returned forthe reporting year 2008. One hundred andninety four ( 67%) of the 290 incidents(SAR/IBCT/ SAE) reported to the NHO met thecriteria for mandatory reporting and werereported on ANSARE. Two blood establishmenterrors were also included on ANSARE.Twenty two sites (29%) reported both SARs andSAEs. A further twenty one (28%) reported onlySARs and six (8%) reported only SAEs. Twentyseven sites (36%) indicated that they had notreported any SAR or SAE during 2008.ANSARE 2009Seventy-five reporting establishmentssubmitted reports in 2009. One hundred andfifty four (58%) of the 267 reports accepted byNHO, were reported on ANSARE. At the timeANSARE was submitted one further caseremained open for further investigation.Sixteen (21%) reporting establishmentsreported both SARs and SAEs. A further 17(23%) reported only SARs and six (8%) reportedonly SAEs. Finally 36 sites (48%) had notreported any SAR or SAE during 2009.CommentThe ANSARE form does not collect nonmandatoryclinical IBCT incidents. These nonmandatoryevents accounted for 32% of thetotal number of reports accepted by the NHOin 2008 and 41% in 2009 compared tomandatory SAEs which made up only 18% in2008 and 19% in 2009. Therefore, theANSARE returns underestimate the overall rateof reporting from reporting establishments tothe NHO.Participation in Haemovigilance 2008 and2009The NHO has examined reporting trendsthrough out 2008 and 2009. Reportingestablishments (RE) are classified in categoriesdepending on the number of componentsissued each year, from the information onANSARE. Seventy one reporting establishmentsissued blood for transfusion in 2008. Threefacilities (hospitals who submitted an ANSAREform) did not transfuse blood components in2008. In 2009, 72 reporting establishmentsissued blood components for transfusion. Thisexcluded two blood establishments and onefacility who did not transfuse bloodcomponents in 2009.Table 5 Reporting EstablishmentCategoriesCategory Components No. of RE No. of REissued issuing issuingblood for blood fortransfusion transfusionin 2008 in 2009(n=71) (n=72)Category A Up to 1000components 34 37Category B 1000 to 3000components 21 19Category C 3000 to 6000components 9 8Category D Above 6000components 7 8In 2008 nineteen reporting establishments didnot submit any mandatory or non mandatoryreports to the NHO. In 2009 twenty tworeporting establishments did not submit anymandatory or non mandatory reports to theNHO. The majority of these reportingestablishments are smaller organisations andfall within the category A group with less than1000 units issued per annum.Figure 1: Reporting establishmentssubmitting between 1 to 5 reports in2008/2009 (n=56)16141210864201316CategoryA98CategoryB36CategoryC2008 200901CategoryD15

In 2008 25 (35%) reporting establishmentssubmitted between 1 to 5 reports (Figure 1).Thirteen (50%) were in category A, 9 (36 %)were category B and three (12%) werecategory C hospitals.Thirty one (43%) hospitals submitted between1 to 5 reports in 2009 a slight increase on 2008figures. Sixteen (52%) were category A, eight(26 %) were category B, six (19%) werecategory C and one (3%) was a category Dreporting establishment.Figure 3: Reporting establishmentssubmitting more than 10 reports in2008/2009 (n=19)654356Figure 2: Reporting establishmentssubmitting 6 to 10 reports in 2008/2009(n=27)1010987632102 21Category B Category C Category D2008 2009654322 23121In 2008 ten reporting establishments (14%)submitted more than 10 reports for review(Figure 3). The majority were as expected fromCategory D reporting establishmentstransfusing greater than 6000 units where fiveof the seven Category D reportingestablishments submitted more than 10reports.10CategoryACategoryBCategoryC2008 2009CategoryDIn 2009 nine reporting establishments (13%)submitted more than 10 reports for review(Figure 6). The majority were again asexpected from Category D reportingestablishments with all six of the eightreporting establishments in Category Dsubmitting more than 15 reports.In 2008 seventeen reporting establishments(24%) submitted 6 to 10 reports (Figure 2).Reporting establishments within category Bcategory submitted 50% (10) of these reports.In 2009 ten reporting establishments (14%)submitted 6 to 10 reports, again the majorityof these reports were submitted from categoryB reporting establishments.CommentReporting of Haemovigilance IBCT/SAE andSAR apart from being mandatory is an integralpart of transfusion safety and is evidence thatthere is active surveillance of transfusion safety.While the reporting pattern of Category A, Band D reporting establishments reflectstransfusion activity in 2008, the reports for thenine Category C (

anges with a third of the hospitals represented ineach report range. This suggests that someCategory C reporting establishments may beunderreporting.In 2009 the overall reporting of SAR/SAE hasdecreased slightly since 2008 and in certainreporting establishments the amount ofcomponents transfused has decreased which mayaccount for this. However it would appear againthat some reporting establishments in category Ccontinue to under report as six of the eightcategory C reporting establishments submittedbetween one to five reports in 2009. Surprisinglyone reporting establishment in category Dsubmitted only 5 reports in 2009 and a secondsubmitted only between 6 to 10 reports comparedto over 15 reports in the other six category Dreporting establishments.Trending Adverse Reaction and Event Reporting2006-2009A review of haemovigilance reporting trends (onblood components including SD plasma) from2006 to date is presented in Figure 7. Reportingon blood products e.g. anti-D and factorconcentrates, both accepted by the NHO, havenot been included in this analysis. Up to 2006, thenumbers of reports analysed by the NHOcontinued to increase year on year since reportingcommenced in 1999.A more detailed analysis of this reporting periodhowever clearly shows while greater numbers ofevents than reactions were reported year on yearup to 2006, this was not the case since 2007.Reports of SAR have overtaken SAE/IBCT by 32and 21 reports respectively.While reporting of non mandatory adverse events(IBCT) occurring in the clinical area reduced byover 50% between 2006 and 2008, this trend isbeginning to reverse in 2009, with 76 reportsrelating to non-mandatory clinical adversetransfusion events having been reported to theNHO, a small increase on 2008 but stillconsiderably below 2006 figures.Key PointsIt is likely that the reduction in reports fromclinical areas from 2006-2008 indicated theemphasis on the requirements of laboratoryaccreditation and mandatory SAE reporting.The slight increase in 2009 non mandatoryreports is encouraging, suggesting this trendis being reversed.Reporting of both mandatory and nonmandatory events make importantcontributions to patient safety and attentionneeds to be refocused on the clinical areas ofblood transfusion not covered by theDirectives.Figure 7: Haemovigilance reporting 2006-2009 (n=993)160 -140 -120 -100 -80 -60 -40 -20 -0 -Total SAR Total SAE Mandatory SAE2006 2007 2008 2009Non Mandatory SAE17

Incorrect Blood Component Transfused(IBCT)/Serious Adverse Event (SAE)2008IntroductionThe NHO collects Serious Adverse Events (SAEs)which are mandatory under legislation (EU BloodDirective 2002/98/EC) and Incorrect ComponentTransfused (IBCT) which are not mandatory butreportable under professional responsibility.The difference between the two definitions is thatthe IBCT category covers errors occurring in theclinical areas of the transfusion chain, such assampling of the patient and administration of thecomponent, whereas SAE cover the quality andsafety of the blood components, focusing on errorsoccurring in the Blood Establishments (BE) and inthe HBB and does not cover errors associated withblood products.An IBCT is defined as:‘The transfusion of a blood component/productwhich did not meet appropriate requirementsand/or was intended for another patient’ (SHOT1996).An SAE is defined as:‘any untoward occurrence associated with thecollection, testing, processing, storage anddistribution of blood and blood components thatmight lead to death or life threatening, disabling orincapacitating conditions for patient or donors orwhich results in, or prolongs hospitalisation ormorbidity’. EU Directive 2002/98/EC andCommission Directive 2005/61/EC.While the NHO to date has not collected near missevents occurring in either the BE or HBB theseevents will be reportable from January 2010.FindingsIn 2008, 147 IBCT/SAE related to bloodcomponents and blood products were accepted bythe NHO, representing 51% of analysed reports.These reports were submitted from 39 reportingestablishments.Thirty six reports related to blood products (factorconcentrates and anti-D) and these are separatelyassessed on pages 39 and 41 respectively.IBCT/SAE associated with blood componentsand SD plasmaThe NHO analysed 111 reports relating to bloodcomponents and SD plasma. Elderly patients agedover 70 years were involved in 25% of reports.Almost 22% of reports involved paediatric patientsunder 18 years of age. Adverse events relating topaediatric patients are covered in detail on page35.Fifty three reports met the criteria of an SAEreportable under EU Directive 2005/61/EC. Theremaining 58 were IBCT due to errors in the clinicalareas.Mandatory SAE reports under Directive2005/61/ECThis is the second year of reporting of mandatoryevents as set out in Commission Directive2005/61/EC. The report is sent to the IMB fortransmission to the European Commission. SAEsare classified by step in the work process e.g.testing of donations, storage, distribution etc. andcause of the adverse event as product defect,equipment failure, human failure or other (Directive2005/61/EC Annex III). These mainly involvedwrongly labelled units or storage issues.In 2008, 53 SAE reports were submitted from HBB.Details of the report submitted to the EC areincluded in Appendix 2.Reporting trends of IBCT/SAEFurther analysis of adverse event reporting clearlyshows that reporting of non–mandatory adverseevents (IBCT) which are primarily events occurringin the clinical area, has substantially reduced. Thenumber of SAE has increased in 2008, formingapproximately 48% of overall adverse events, whilereports of non-mandatory IBCT has continued to18

decrease. Only 52% of reports of adverse eventsrelating to blood components and SD plasma wereIBCT, compared to 66% and 79% in 2007 and 2006respectively.The changing profile of IBCT/SAE reporting isillustrated in IBCT/SAE Figure 1. While the numbersof ABO/Rhesus (Rh)D SAE have remained relativelyconstant, numbers of Unnecessary Transfusionshave also reduced since 2007 and again in 2008,with reduced reports of IBCT.IBCT/SAE Figure 1: Changing profile of IBCT reports 2000-20082001801601401201008060402002000 2001 2002 2003 2004 2005 2006 2007 2008Total SAE/IBCTWrong ABO/Rh D group/Wrong Component/Blood to Wrong PatientUnneccessary TransfusionCategorisation of IBCT/SAE analysed byNational Haemovigilance Office –Whathappened?All reports analysed by the NHO wereinitially categorised by the nature of theadverse event experienced by the patient.This maintains the clinical focus of thereporting system.19

IBCT/SAE Figure 2: Reports analysed by the NHO (n=111)Blood to wrong patientOver transfusionIncorrect Rh D group transfusedIncorrect ABO group transfusedTransfusion of expired componentTransfusion of other antigen incompatible RCCFailure to give CMV negative, irradiated componentIncorrect component transfusedUnnecessary transfusionTransfusion of incorrectly stored componentOtherTransfusion of an incorrectly labelled unit------------| | | | | | | | | |0 2 4 6 8 10 12 14 16 18 20MandatoryNon MandatoryMajor clinical findings• There were fifteen reported cases ofunnecessary transfusion, nine cases involvingred cells, four reports involving SD plasma, onereport involving platelets, and one reportinvolving an unnecessary transfusion of bothred cells and platelets.• Thirteen reports involved the transfusion of theincorrect components/product to patients. Fivecases involved red cells, five involved SDplasma, two cases involved platelets and onecase involved fresh frozen plasma (FFP).• Four patients received components which werethe incorrect ABO group. One report involvedred cells, two involved platelets and one SDplasma. There were no reactions.• Four patients were transfused with blood of theincorrect RhD group.• There were seven reports of patients notreceiving cytomegalovirus (CMV) negativeand/or irradiated blood.• There were six reports of patients not receivingantigen compatible red cells.• The wrong blood was given to a neonate(IBCT/SAE Case 12).• There were three cases where a neonate andinfants were over-transfused (IBCT/SAE Case17).• There were 17 reports captured as “Other”.One case involved a failure to providephenotyped units for a patient in sickle cellcrisis (See Appendix 3).20

Adverse events associated with errors atlabelling, storage, administration and othermiscellaneous errors• Nineteen cases were reported where thecomponents were incorrectly labelled. Redcells were involved in 15 cases and SD plasmain two cases. Platelets and granulocytes wereeach involved in one case. The majority ofthese were classified as SAE.• There were 17 reports of transfusion ofincorrectly stored components, 11 involvingred cells and three involving both SD plasmaand cryoprecipitate.• There were five reports of transfusion ofexpired components, all involving red cells.Four cases were considered SAE involvingissues from the HBB and in the final case, aclinical decision was made to transfuse anexpired unit of red cells in theatre.• The remaining reports captured as “Other”included the following adverse events; use ofincorrect giving set (3), transfusion exceeding 6hours (7), pack perforated during transfusion(2), transfusion of uncross-matched red cells (1),clots reported in pack (1) (See Appendix II).Unnecessary transfusion (n=15)The NHO received 15 reports in this category in2008. This was a reduction of 11 reports onprevious year. Nine reports involved red cells, fourSD plasma, one involved platelets and one,multiple components (red cells and platelets).While reports of unnecessary transfusions inpatients of all ages were received, 47% of thepatients involved were over 70 years of age.Most of these events occurred at theprescription/request step of the work process(n=13), one involved wrong results from thehaematology laboratory and the final eventoccurred because of delay in bringing the patientfor the procedure.An overall review of these reports revealedunnecessary transfusions occurred as a result of:• Decision making which deviated from clinicalguidelines (n=10)• Decision making based on incorrect or absenthaematology results (n=4)• Delay in transfusion (n=1)Decision making which deviated from goodclinical practice (n=10)Seven cases involved red cells; five of which wereunnecessary transfusions administered to patientswith iron deficiency anaemia. A further caseinvolved a patient with megaloblastic anaemia dueto vitamin B12 deficiency (IBCT/SAE Cases 1 and2). The final case of over-transfusion in the contextof haemorrhage is described in IBCT/SAE Case 3.All of these patients received these transfusions asresults of errors made by medical staff prescribingblood components.Unnecessary red cell transfusionsTransfusions for iron deficiency anaemiaTwo patients had iron deficiency anaemia as resultof menorrhagia. One of these is described below(IBCT/SAE Case 1).IBCT/SAE Case History 1A 43 year old female patient developed anurticarial reaction post transfusion of red cells.On investigation, the HVO discovered thepatient had a haemoglobin (Hb) of 7g/dl, dueto underlying menorrhagia. She wasasymptomatic and had not been commencedon iron. The cause of this error was lack ofknowledge of the prescribing doctor.In another case an elderly patient with irondeficiency anaemia received an unnecessary redcell transfusion, the cause of error was identified asa lack of knowledge by the prescribing doctor andabsence of guidelines on management of patientswith chronic anaemia.There were three further cases where patients withchronic iron deficiency anaemia receivedunnecessary red cell transfusions. In one of thesecases, the patient had been reviewed by thehaematology team who had ordered intravenous(IV) iron for asymptomatic anaemia. It was unclearwhy this advice was not followed.21

22Transfusion for megaloblastic anaemia due tovitamin B12 deficiencyIBCT/SAE Case History 2This young male adult patient was admitted viaAccident and Emergency (A&E) from GeneralPractitioner (GP) for investigation of anaemiawith a differential diagnosis of B12 or folatedeficiency. While this patient had a history offatigue, he did not have dyspnoea or anyhistory of haemorrhage. His Hb was reportedat 6.3g/dl. Two units of red cells wereprescribed by a junior hospital doctor workingin the (ED) emergency department. Thisoccurred outside routine working hours. Oneunit was transfused. When this patient wasreviewed by the haematology team, thesecond unit of red cells was cancelled.Unnecessary transfusion for haemorrhageIBCT/SAE Case History 3In the final case involving unnecessary red celltransfusion, a patient admitted with a Hb of13.7g/dl and PR bleeding developed atachycardia and mild hypotension 24 hours postadmission. While there was further bleeding, thepatient received five units of red cells withoutchecking between units and had a Hb of 15.2g/dlpost transfusion. Some of the units wereunnecessary.Three cases are reported in the paediatric section,where patients were over transfused (See IBCT/SAECase 17).Unnecessary SD Plasma and PlateletsTwo patients received unnecessary SD plasmatransfusion caused by lack of hospital policies andknowledge of prescribing clinicians.One patient on warfarin with an INR of 3.5 wastransfused platelets instead of Vitamin K and PCCprior to insertion of a chest drain. This eventoccurred when the platelets were prescribed by aconsultant who did not prescribe bloodcomponents routinely.Key Points• Transfusions for anaemia due tohaematinic deficiency accounted for 40%of all unnecessary transfusions and 5% ofall IBCT/SAE reported in 2008.• Asymptomatic patients with irondeficiency anaemia should be treatedwith iron therapy. Oral iron should becontinued for at least three months afterdeficiency has been corrected so that ironstores are replenished. Ascorbic acid mayenhance iron absorption. Where there areconcerns about compliance, IV ironproducts should be considered. (Goddardet al, 2005)• Patients with megloblastic anaemiarespond very rapidly to vitamin B12 andfolate and very rarely require transfusion.• Where several units are prescribed fortransfusion, the patient's Hb should bechecked between units. This will minimiserisk of unnecessary/over transfusion(NBUG, 2001).Decision making based on incorrect or absenthaematology results (n=4)These decisions resulted in three unnecessary redcell and one platelet and red cell transfusion. Intwo cases, patients received unnecessary red cellswhen doctors failed to verify blood results.In another case, a doctor ordered three units of redcells over three consecutive days based on Hb fromthe initial day. No Hb checks were done betweenunits. The final event is described in the paediatricsection (IBCT/SAE Case 16).Unnecessary transfusion because of a delay(n=1)IBCT/SAE Case History 4This case involved the unnecessary transfusionof three units of SD plasma. A patient having aliver biopsy under radiological guidance wastransfused SD plasma on the ward for anabnormal INR. Although the consultantradiologist requested that the patient attendthe X-ray department at a certain time, this wasdelayed by the plasma transfusion. When hewas transferred to the X-ray Department, it wastoo late to undergo the procedure as it ishospital policy that all patients have consultantdelivered care for two hours post procedure,and this was unavailable because of the delay.

Incorrect ABO group, RhD group, wrongpatient and antigen negative blood transfused(n=15)There were four cases involving transfusion of acomponent of the wrong ABO group, fourinvolving errors in Rh group, one event where theincorrect patient was transfused and six involvingfailure to transfuse antigen negative blood.Incorrect ABO group transfused (n=4)The NHO received four reports in this category.Although only one was associated with red cells,these cases were considered to have highpotential to cause harm. However, there were noreports of SAR associated with ABOincompatibility in 2008.IBCT/SAE Table 1: Age and Componentimplicated in transfusion of ABO incompatibleunits (n=4)Component Neonate Infant Adult(

• Group O platelets, unless they aresuspended in platelet additive solution,particularly O apheresis platelets,should be reserved for group Opatients.• Transfusion of group O apheresisplatelets to patients whose bloodgroup is A, B or AB has potential tocause haemolysis. (BCSH, 2003; NHO,2007; SHOT, 2009).Incorrect RhD group transfused (n=4)The NHO received four reports in this category.All reports in this category involved red cells.Fortunately three of the four RhD positive redcells were transfused to male patients and thefourth was transfused to a post menopausalfemale.All errors occurred in the HBB. Two errors wereincorrect selection of RhD group, in another casea computer warning/flag was over-ridden andthe final error resulted from incorrecttranscription of results (IBCT/SAE Case 7).IBCT/SAE Case History 7The medical scientist working in the hospitalbank carried out a manual group andcrossmatch and recorded the resultincorrectly. While the patient’s blood groupwas correctly typed as O RhD negative, it wasrecorded as O RhD positive. This was anemergency crossmatch and, as this was out ofhours, the automated grouper was not used,as training in its use had not been fully rolledout to all on-call staff. This error occurredwhen one medical scientist was on cross-callcover in the HBB. The error was discoveredby another medical scientist at the nextcrossmatch.Error cause: The root cause identified was asimple human lapse in concentration, where amedical scientist incorrectly transcribedresults. That the automated grouper was notused outside routine hours most likelycontributed to the error.An overview of root causes across the remainingthree cases identified one other system failure.In this case, a system error was reported when amedical scientist working in the HBB, over-rode acomputer flag and issued RhD positive red cellsto a patient whose blood group was RhDnegative. It was usual practice for patients in thishospital to be medically reviewed late in the day,and therefore cross matching was done out ofhours. This was reported as an organisationalsystems failure involving hospital culture whichcontributed to less safe practice.Human failures also contributed to patientsbeing transfused with incorrect RhD group redcells. These included human slips, lack ofknowledge, a failure to follow policies, coordinationand communication.In one case, where group RhD negativecomponents were unavailable, the medicalscientist selected RhD positive componentswithout informing the Consultant Haematologist.This was not an emergency transfusion and wasprobably inappropriate, as the patient was beingtreated for iron deficiency anaemia. Had the casebeen reviewed medically, this transfusion wouldnot have proceeded.There were contributing factors reported in threecases.Three adverse events occurred out of hours andin two of the three cases the medical scientist didnot routinely work in the HBB. One case was anemergency crossmatch.Corrective and preventative action was reportedin three cases. This included the introduction of acomputer flag to minimise the potential forissuing RhD incompatible components, a memoto all staff working in the HBB on componentselection, a feasibility review of out of hours useof the automated grouper and a plan to reviewout of hours crossmatch and transfusions whererequests were routinely received late in theworking day.Wrong blood given to a patient (n=1)This case is described in the paediatric section(IBCT/SAE Case 14).24

Transfusion of antigen incompatible red cells(n=6)The NHO received six reports in this category.None of the patients had an associatedtransfusion reaction. One male patient was only15 years of age and is described in the PaediatricSection (IBCT/SAE Case 15).IBCT/SAE Table 2: Transfusion of antigenincompatible red cells (n=6)Age Gender Patientantibody statusAdult 31-50 years Female Anti-Jk a antibodyAdolescent 12-17 years Male Anti-Jk a antibodyAdult 18-30 years Male Anti-Kell antibodyAdult 31-50 years Female Anti-Kell antibodyElderly 70 years Male Anti-Fy a antibodyAdult 51-70 years Female Anti E (enzyme only)In two cases, antibodies although present werenot detected on the pre-transfusion crossmatch.In one case, the pre–transfusion test lackedsensitivity to detect Anti-Jk a antibodies. Althoughit can be particularly difficult to detect Jk aantibodies, in this case it was detected duringretrospective testing of the pre-transfusionsample when a different test was employed. Thelack of a sensitive pre-transfusion test wasidentified as the root cause of these events. Thiswas classified as a system failure, materials.In the second case, an Anti-Fy a antibody notdetected prior to transfusion was subsequentlydetected on re-testing of the pre-transfusionsample. The reason for this was unclear.In two cases, removal of a computer flag left staffunaware that a patient required antigen negativered cells. In another case, a female patient ofchild-bearing potential received a Kell positiveunit of red cells. This occurred when a medicalscientist failed to carry out a final check to ensurethe patient’s age and the Laboratory InformationSystem (LIS) was not set-up to prompt this. In thefinal case, the medical scientist entered theincorrect result (IBCT/SAE Case 8).IBCT/SAE Case History 8In this case, a medical scientist processing aroutine sample during on call hoursincorrectly entered the results of the crossmatchas negative, when it was positive. Anenzyme only anti-E was subsequentlydetected in the patient sample. Althoughenzyme only anti-E is unlikely to be clinicallysignificant, the potential for harm of this errorwas high.Error Cause: The root cause of this errorwas identified as a design systems failure.Had the blood group analyser beeninterfaced with the LIS, this error would nothave occurred. Human error was alsoinvolved in this SAE as the medical scientistwas working on-call and did not routinelywork in the HBB. If the request had beenprocessed during routine hours, thelikelihood of this error occurring would havebeen reduced.All of these events were discovered in the HBB,three at the next cross-match, two posttransfusion either at the time of fating of units orauthorisation of results, and one event wasdiscovered when an analyser was undergoingvalidation.Key Points• Patients who have had previouspregnancies or transfusions are at risk ofdeveloping antibodies. Very often theHBB can be unaware of patients’ historyand the potential for antigenincompatible transfusions can be high.Development of a national antibodyregister could address this risk byensuring access to patients’ antibodyhistory. It would also reduce therequirement for repeat laboratorytesting.• This would only be feasible with theimplementation of a national uniquehealth identifier (UHI), arecommendation made by the Healthand Information Quality Authority(HIQA) (2009) and supported by theNHO. A UHI would facilitate improvedand safer access to patients’ records ona national antibody register therebyensuring safer transfusion practice forpatients.25

26Incorrect component transfused (n=13)The NHO received 13 reports in this category,which captures reports of patients who received anincorrect component, when another componentwould have been more appropriate. Of the 13cases reported, five involved infants and these areseparately discussed in the paediatric chapter.Six cases involved plasma (SD plasma n=5, FFPn=1), five involved red cells and two platelets.IBCT/SAE Case History 9An adult patient received a unit of uncrossmatchedred cells. A medical scientist on callreceived a sample for grouping but crossmatchedtwo units for a patient in A&E whowas likely to require the blood. At this time, thecrossmatch had not been requested, so themedical scientist did not label or issue the unitsto the patient. This medical scientistcompleted the on-call shift. When acrossmatch was later requested post midnight,a second on-call medical scientist labelled andissued two units of red cells, thinking they werethe already crossmatched units. These unitswere transfused, but one of these units had notbeen previously crossmatched, as the medicalscientist had selected the incorrect unit.Error Cause: Systems failure-culture wasidentified as root cause of the error. Thepractice of anticipating the need for blood andcross matching red cells to assist colleagueswithout labelling them as cross-matched, wascommon practice in the HBB.• There were five cases involving transfusion ofSD plasma to reverse warfarin whenprothrombin complex concentrate (PCC) wouldhave been the correct product. All of theseerrors occurred at prescription/request andinvolved medical staff.• One patient received FFP instead of SDplasma. This error occurred in a hospital whereboth FFP and SD plasma were stocked in theHBB. Clinical staff commonly referred to SDPlasma as FFP when requesting plasma soconfusion arose when selecting the correctplasma.• Two cases were reported where the incorrectplatelet product was transfused. One caseinvolved an adult patient who should havebeen transfused with Human LeucocyteAntigen (HLA) matched platelets (IBCT/SAECase History 10). The second case involved aninfant who received a pooled instead ofapheresis platelets.IBCT/SAE Case History 10A patient requiring HLA matched platelets wastransfused two units of random platelets. In thiscase, the patient’s consultant wanted HLAmatched platelets for the patient to cover aprocedure. A clinical nurse specialist caring forthe patient ordered the HLA matched plateletsfrom the BE some days ahead of theprocedure, but did not inform the HBB.Platelets were issued from the HBB on thebasis of this prescription which did not specifythe requirement for HLA matched platelets.Site of ErrorFirst site of error was reported asprescription/request in 62% (8) of these casesand laboratory processing-blood transfusion in31% (4) cases. One error was made atadministration.An overview of causal factors contributing toprescription/request errors identified primarilyhuman failures such as lack of knowledge (6 cases),failure to adhere to policies (2 cases), failure ofcommunication or co-ordination of care betweendisciplines (4 cases).A review of these errors involving the HBBidentified system failures in one case and humanfailures in all cases.Transfusion of incorrectly labelled units (n=19)The NHO received 19 reports in this category.Fourteen reports were mandatory SAE where theerror occurred in the HBB. In five cases the erroroccurred in the clinical area e.g. at initial clerking,sampling, collection and administration. In fourreports, it was noted it was an emergencytransfusion and in five cases the adverse eventsoccurred out of hours.Sixteen reports in this category involved red cells.platelets, SD plasma and granulocytes were eachimplicated in one report. Sixteen events involvedadult patients and three paediatric patients. Theerrors were classified as follows:

• Transposition of labels within a single crossmatch (n=10)• Incorrect data on label- patient identifiers, unitnumber (n=6)• Unlabelled units transfused (n=3)IBCT/SAE Case History 11In this case, unlabelled and uncrossmatchedgranulocytes were transfused to a patient. Thecomponent was ordered by the patient’sprimary care team from the BE, however, thisrequest was not communicated to the staff inthe HBB. This HBB had not had an order forgranulocytes in a very long time and wereaware that the haematologists in the hospitalhad never used granulocytes. They wereunaware that on this occasion they wererequired for a patient in the hospital, and infact, they believed that the granulocytes wereintended for research. Therefore, the HBB didnot accept the granulocytes, which weredistributed directly to the patient in the ward,where a clinical decision was made to transfusewithout crossmatch.Error cause: Had the prescribing doctorcommunicated his request to the HBB, thisadverse event would have been averted. TheHBB failed to communicate with the clinicalteam or the BE prior to declining thegranulocytes. The HBB did not have anypolicies to manage receipt, crossmatching andissuing of granulocytes, as they did notroutinely use these components in the hospital.Transfusion of incorrectly stored units (n=17)The NHO received 17 reports where patientsreceived blood components or SD plasma whichwas incorrectly stored. Reports were analysed asfollows:IBCT/SAE Table 3: Analysis of reports ofincorrectly stored units transfused(n=17)Analysis of incorrectly n Mandatorystored unitsSAEUnits returned to controlled 5 4storage after 30 minutes,subsequently removed andtransfused greater than fourhours after initial removalfrom controlled storageAnalysis of incorrectly n Mandatorystored unitsSAEUnits stored in uncontrolled 5 1storageProblems with controlled 5 5storageOther 2 1Where reports involved either clinical personnel orSD plasma, these were captured as non-mandatoryreports as outside the Commission Directive.Units returned to controlled storage after 30minutes, subsequently removed and transfusedgreater than four hours after initial removal fromcontrolled storage (n=5)Red cells should be transfused to intendedrecipients within four hours of removal fromcontrolled storage.There were five reports in this category, all involvingsingle unit transfusions. In each case, the unit of redcells was not returned from the clinical area tocontrolled storage within 30 minutes and wassubsequently removed and transfused to theintended patient over four hours after initialremoval from controlled storage.These events occurred because the blood wasbrought to the bedside without checking thepatient’s status - two patients were pyrexial, onepatient had incorrect information on his identity (ID)band, one patient had been moved to another unitand there were difficulties in gaining intravenousaccess.These adverse events were primarily caused byhuman errors: failure to adhere to policies, lack ofverification, lack of knowledge, task carried outincorrectly.Key Points• Best practice guidelines indicate that apatient’s status should be checked prior tobringing a unit of blood to the bed-side(BCSH, 1999, NBUG, 2004). Adherence tothese guidelines will reduce such reportsand minimise the risk of adverse events27

and potential wastage of a scarceresource.• If transfusion is clinically indicated, itshould not be delayed solely becauseof pyrexia. Therefore, transfusions canbe commenced on pyrexial patients,but these patients should be verycarefully monitored throughout thetransfusion.Units stored in uncontrolled storage (n=5)There were five reports in this category. Twoinvolved red cells, two cryoprecipitate, and oneSD plasma.Two units of red cells were stored in uncontrolledrefrigerators. In the first case, a unit of red cellsincorrectly scanned from the blood bank using anelectronic blood-tracking system wassubsequently placed in an uncontrolled fridgewhen the satellite fridge could not be accessed.This was done on instruction of the medicalscientist in the HBB as the blood-tracking systemdid not recognise the unit of red cells. Thissystem had just been introduced into the hospitaland not all staff were trained in its use.In the second case, a nurse on night duty placeda unit of red cells in an uncontrolled fridge in award area. She had not checked if the patient’s IVcanulae was correctly placed prior to orderingthe blood. This was her first week of night duty ina new hospital, and she had not received formaltraining.Three adverse events involved storage ofcryoprecipitate and SD plasma in satellite fridges.Problems with controlled storage in the HBB(n=5)All of these reports involved multiple units. Threecases involved red cells and two cases involvedSD plasma and cryoprecipitate.In three cases, red cells were stored in fridgeswhich had not been validated by temperaturemapping, or where temperature range was notcontrolled due to an equipment failure. All ofthese errors were reported as mandatory SAE.In two cases, cryoprecipitate and SD plasma werestored in freezers where the temperature wasoutside the correct range.Key Point• These cases highlight the importance ofboth initial haemovigilance orientationand ongoing training for clinical andlaboratory staff in storage and handlingof blood components.Other (n=2)These reports involved multiple units of red cellsand SD plasma. In the first case, red cellspreviously identified as been incorrectly storedwere returned to controlled storage, reissued andtransfused. In the latter case, units of SD plasmapost de-frosting were stored in the clinical areaprior to transfusion.Discovery Information - Who discovered theerror?This year, unlike previous years, 38% of adverseevents were discovered by medical scientists,who discovered 55% of mandatory errorsoccurring in the HBB.HVOs discovered 33% of all events and 48% ofnon-mandatory clinical IBCT.Nurses discovered 17 (15%) adverse eventsfollowing commencement or completion of atransfusion. Eight adverse events werediscovered and reported by doctors, generallyduring clinical review of patients. These weremainly clinical (non-mandatory) errors. One eventwas discovered by an Inspector during theinspection process. This was a mandatory SAE.IBCT/SAE Table 4 Who discovered the adverseevent? (n=111)Who n Mandatory NonMandatoryMedical Scientist 42 29 13HVO 37 9 28Medical Scientistand HVO 6 4 2Nurse 17 9 8Doctor 1 8 1 7Other 1 1 0Total 111 53 581Nurse and Medical Scientist also aware of this error.28

At what point in the work process wereadverse events discovered?Seventy two (65%) adverse events werediscovered following the transfusion.Medical scientists discovered 42 events duringroutine laboratory activity. The majority (34) werediscovered at;• Next crossmatch n=8• Next issue within same transfusion event n=5• Fating of units transfused n=5• Post call check n=8• Stock reconciliation n=8Unnecessary transfusions were discovered byHVO (7), medical staff (4) and medical scientists (4)following transfusion.Key Point• This is the first year that HVOs did notdiscover the majority of reportedadverse events. This probably reflectsimproved vigilance systems, either interms of surveillance or reporting in thelaboratories associated withInternational Standards Organisation(ISO)15189.• Yet again in 2008, a majority of errorswere discovered during post transfusionsurveillance activity. This indicates thatpotentially many errors could have beendetected by staff in the transfusionprocess prior to completion oftransfusion, highlighting the need forcontinuing education for clinical andlaboratory staff involved in thetransfusion process.• Error discovery by doctors increased in2008. While the numbers are still small,this is encouraging, indicating anincreased awareness of the importanceof reporting adverse transfusion events.Error occurrence-where did the error occur?The site of first error in the work process wherethe adverse event occurred and the discipline ofpersonnel involved is illustrated in IBCT/SAE Table5.IBCT/SAE Table 5: Site in transfusion process where error first occurred & discipline involvedin IBCT/SAE (n=111)Stage of Work Process Total Discipline involvedn Nurse Doctor Medical Scientist Porter OtherBE 1 0 0 1 NA NAInitial Clerking 1 NA NA NA NA 1Sampling 1 0 1 NA NA NAPrescription Request 29 5 29 2 1 NA NALaboratory (Other) 1 0 NA 1 NA NAHBB 40 15 2 40 3 2 NAStorage 11 6 0 6 4 0 NACollection 6 4 0 0 2 NAAdministration 18 15 5 3 NA NA NAOther 3 1 1 1 0 NATotal 1112Nurses were also involved in five errors along with medical staff, and the HBB was also involved in one adverse event.3All events occurring in HBB were not identified by nurses, doctors and porters, therefore they are also implicated inadverse event.4One event involved staff in the blood transfusion laboratory and a nurse.5One event involved a nurse and a doctor.29

Key Points• A comparison of data between 2007and 2008 shows thatprescription/request, bloodtransfusion laboratory processing andadministration continue to be thesteps in the work process where mostadverse events begin.• However in 2008, reports associatedwith blood transfusion laboratoryprocessing surpassedprescription/request for the first timesince reporting commenced inIreland, possibly reflecting the impactof quality systems in the HBB.Risk assessmentA risk assessment based on potential impact ofharm and frequency of events. High riskactivities can be identified by examining eachstep of the transfusion work process in terms ofadverse event occurrence and examining thenumbers of events with high potential to causeharm to patients.IBCT/SAE Table 6 presents data on steps of thework process; number of adverse events, errorincidence reported in 2008 and percentage ofevents with high potential for harm for patients.IBCT/SAE Table 6: Risk Assessment by stepsof the blood transfusion process. (n=104)Step in Total events % of events perwork at step in step in the workprocess work process process withhigh potentialfor harmPrescription/Request 29 62%BloodTransfusionLaboratory 40 52%Storage 11 27%More events with greater potential for harm topatients occur at Prescription/Request and in theblood transfusion laboratory. This includesevents such as unnecessary transfusion, incorrectcomponent/product transfused, incorrect ABOand RhD group transfused.While the events reported with first site of errorat administration did not have high potential tocause harm, the bedside check is the finalbarrier to identify and prevent adverse events(Reason, 2000). While administration practiceappears safer than other areas, it is clear fromreports examined in 2008 that this barrier maynot be optimally functioning. Reported adverseevents such as transfusion of incorrect ABO/RhDgroups, units with transposed labels andunnecessary transfusions were not recognised atthe bedside. As the NHO does not acceptreports of clinical near miss events, the extent towhich the bedside check is working is not clear.It is important to emphasise that the bedsidecheck is crucial to identify and preventtransfusion adverse events.Key Points• Based on analysis of data received in2008, while all steps in the bloodtransfusion process have potential tocause harm to patients,prescription/request and bloodtransfusion laboratory processing arepotentially more risky to patients.• This analysis highlights areas of thework process where haemovigilanceresources should be concentrated, witha clear objective of enhancing patientsafety i.e. on prescribing/requesting,and HBB practices.• The role of the bedside check as abarrier to unsafe transfusion practicemust be highlighted to clinical staff,nurses, medical staff and perfusionistsin haemovigilance education sessions.Collection 6 33%Administration 18 22%30

Overview of changes to practiceFollowing investigation and evaluation, thepurpose of follow-up action is to address the rootcause of the event. An analysis of reports receivedin 2008 indicated 50% (55) implementedcorrective action, 41% (45) did not implementcorrective action and in 10% (11) , this informationwas not provided.The requirements of the Directive 2005/61/ECrequire reports of corrective action following SAE.Corrective action was specified in 51% (27) reportsof SAE. A review of reports where corrective actionwas implemented did not identify any trend in thisarea. Corrective action was implemented across allcategories of reports, both mandatory and nonmandatory, both high and moderate risk.Reported follow–up action can be categorised asfollows.IBCT/SAE Table 7: Follow-up action forIBCT/SAE (n=55)Category offollow–up action Details nProcess changesEducation andTrainingCommunicationLaboratory practiceschanged.Primarily an introductionof a further check 17GeneralhaemovigilancetrainingTargeted updates.Individual retrainingGeneral retrainingE Learning 15Memos to staffDevelopment of postersRaising awareness 8Development and Development of newrevision of policies policiesRevision/change tocurrent policies 6Audit Current clinical practice 2Multiple follow–up Including development/actionsrevision of policies,education,communication, clinicalpatient review etc. 6Other Order new equipment 1Key Points• It is encouraging that reported follow-upaction has increased from 37% in 2006 to50% in 2007 of all cases analysed byNHO.• The NHO recommends a systematicapproach to follow-up action in terms ofplan development and evaluation ofchanges to evaluate their impact onpractice.Overview of causal analysis –Root cause ofeventsAn analysis of reports where adverse eventsoccurred at prescription/request, HBB, storage,collection and administration revealed multiplecontributing causes.Human ErrorUp to 167 human errors contributed to 104 events.The most commonly reported human errors werefailure to adhere to policies (51) and lack ofknowledge (despite training) (33) and carrying outtasks incorrectly (20). A detailed description ofeach classification of human error can be found inAppendix 4. These errors caused adverse eventsto occur at all steps of the work process reviewed.IBCT/SAE Figure 3: Human errors leading toevents (n=167)6050403020100-------Failure toadhereto policiesKnowledgeCarrying outtask incorrectlyVerificationCo-ordination/communicationThe most frequently reported cause of humanerror in both clinical IBCT and SAE in the HBB wasa failure to adhere to policies. Lack of knowledgewas the second most common contributing factorto clinical IBCT. The third most commoncontributing factor wascommunication/coordination breakdown,reflecting the complex nature of clinical practiceSlipMonitoringUnclassifablePatient related31

with multiple teams and professionals caring forpatients receiving transfusions. Lapses inconcentration (slip) and failure to verify practicefrequently contributed to reported SAE in the HBB.IBCT/SAE Table 8: Recurring human error inclinical IBCT and SAE in HBBClinical IBCTSAE in the HBB1 Failure to adhere Failure to adhere toto policiespolicies2 Knowledge Slip3 Co-ordination/CommunicationVerificationSystem ErrorUp to 40 system errors contributed to 36 events. Nosystem failures were reported in the remaining 68events. The most commonly reported system errorswere lack of policy (11), design of process/systems(10) and management priorities (7). A detaileddescription of each classification of system error canbe found in Appendix 4.IBCT/SAE Figure 4: System errors leading toevents (n=40)121086420------No policyDesignManagementPrioritiesAllA comparison of the most frequently reported causesof system failure identified similar contributing factorsto adverse events in both clinical area and in theHBB.MaterialsCultureOtherIBCT/SAE Table 9: Recurring system failure inclinical IBCT and SAE in HBBClinical IBCTSAE in the HBB1 Unclear or absent policy Unclear or absentpolicy/Inadequatesystem-processdesign2 Inadequate process design3 Failure of management to Failure ofprioritise safetymanagement toprioritise safetyAnalysis of findingsExamining frequently occurring contributing factorsenables identification of factors common to mostevents. From reports received this year, humanfailures, failure to adhere to policies/procedures, lackof knowledge, system failures and lack ofpolicies/procedures governing processes and designwere the most frequently reported causes of events.1. HUMAN FACTOR - LACK OF KNOWLEDGE-TRAINING AND COMPETENCESimilar to NHO report (2007), the failure ofknowledge although staff had received training, wasthe second most frequently reported human error thisyear, especially for clinical staff implicated in IBCT.a. Medical StaffPoor knowledge as a contributing factor totransfusion error is not new. Previous NHO reportshave also identified this, and it is also comparablewith studies on medication error (Leape at al, 1995).One option is to comprehensively address thisdeficiency during training, however, there is currentlyno agreed curriculum in transfusion at undergraduatelevel in medical schools in Ireland.This leaves a gap which hospitals try to fill byproviding appropriate and focused training inhaemovigilance and blood transfusion to staffinvolved in transfusion, which is delivered by HVOs,medical scientists and haematologists. Althoughevidence of training of staff in blood transfusion ispart of the INAB’s interpretation of ISO15189 atinspections, in many hospitals, there is still pooruptake of these sessions by medical staff especially atSenior House Officer (SHO), Registrar and Consultantlevel.32

Other options such as collaboration between theacademic institutions and hospitals should beexplored. A recent study carried out at Beaumonthospital and the Royal College of Surgeons inIreland (RCSI) (Robb et al 2009) showed increasedconfidence among medical students inundertaking clinical and practical skills followingcompletion of an intensive two week “subinternship”programme. During the two weekduration of this programme, final year medicalstudents assumed the role of “highly-supervised”interns, participating in clinical rounds, caseconferences and event on-call schedules.Inclusion of transfusion at this point may focus onclinical transfusion skills of future interns workingin hospitals.b. Medical ScientistsThe requirement for training and competencyassessment for all staff working in the HBBincluding out of hours staff has been clearlydefined by both ISO15189 and AML-BBMinimum Requirements for Blood BankCompliance with Article 14 (Traceability) andArticle 15 (Notification of SAR/SAE) of EUDirective 2002/98/EC and is audited duringinspections.More recently specific recommendationsregarding training and competence for medicalscientists working in the HBB have been made bythe UK Transfusion Laboratory Collaborative(Chaffe et al, 2009). This collaborative is made upof both scientific and medical expertise andrecent recommendations arose following analysisof adverse events reported to the UK SHOTscheme. The recommendations include:• minimum qualifications for medical scientistsin HBB,• ongoing training and competency assessmentfor core and on call medical scientists in theHBB,• availability of specialist scientific adviceduring on call hours.These recommendations have considerableresource implications.c. Competency AssessmentCompetency is a process which denotes not onlyknowledge, but also a technical proficiency(Office for Health Management, 2004). In the UK,the National Patient Safety Agency (NPSA)launched competencies in defined aspects of thetransfusion work process e.g. obtaining a venousblood sample, organising the receipt of blood fortransfusion etc. (NPSA, 2006).A recent pilot study in Scotland found bothclinical assessors and practitioner’s experience ofa formalised clinical competency onadministration of blood components to be verypositive (Pirie and Gray, 2007). This small studyfocussed on nurses and did not examine theperceptions of other staff in the transfusionprocess e.g. portering, medical scientists ormedical staff.Implementation of the NPSA standards has variedthroughout the UK but they have been fullyimplemented in Northern Ireland (NI) with thesupport of the NI Regional TransfusionCommittee. An evaluation of clinical practicefollowing implementation of these standardswould be useful.Competency assessment for all clinical staff maybe challenging especially to ensure engagementby multidisciplinary teams. It is also resourceintensive and requires organisational support forthe initiative at senior clinical and managementlevels.d. Role of hospitalsAs detailed above, training programmesprovided in hospitals are important in improvingtransfusion safety.There is, however, a significant body of evidenceshowing that effective learning requires not onlyappropriate training, but also an organisationalculture which reinforces this training (Senge1990). Culture change can be internally led bylocal champions who support good practice. Thismay be the Consultant Haematologist, the HVO,the medical scientist or the anaesthetist in theatresupported by the hospital transfusion committee.Hospital management must also support safetransfusion practice by developing a learningculture through ensuring protected time foreducation, accessibility to online transfusionresources and support for reporting and learningfrom adverse events.33

e. Role of NHOThe NHO continues to deliver formal educationopportunities in collaboration with DCUtargeting primarily clinical and laboratoryhaemovigilance staff.Since 2007, the IBTS/NHO in collaboration withthe SNBTS actively encouraged hospitals tomake the web based transfusion elearning/competency assessment programmeLearnpro NHS available to their staff.2. System Factor-Policy DevelopmentWhile the initial response to an adverse eventmay be to develop new policies, in many casesfailure to adhere to already existing policies wasevident in both the clinical area and in the HBB.The NHO has reported similar findings both in2007 and 2008. Political, organisational,scientific and cultural concerns ranging frominadequate staffing levels, busy rosters,concerns about limitations on clinical judgementall act as barriers to successful implementationof policies (The Appraisal of Guidelines Researchand Evaluation (AGREE) Instrument, 2003; Boseet al, 2006; Napoli and Jagoda, 2007). Staffdeveloping clinical and laboratory policiesshould be aware of these concerns.3. System Factor- Process/System DesignPoor design of processes and systems was thesecond most frequently reported system error in2008. This emphasises the importance ofevaluation of all aspects of systems and theirimpact on practice, especially where changeshave been introduced.Key Points• Training of staff especially medical staffis complex. Implementation of astandardised transfusion programmefor undergraduate students shouldensure emerging clinicians will have anunderstanding of safe transfusionpractice. The role of the practiceeducation delivered by the HVO,Consultant Haematologist and seniormedical scientists is also criticallyimportant to integrate theory and safepractice.• Initiatives such as audit, provision offeedback, presentations or acting as a“clinical/laboratory” presence by theConsultant Haematologist, hospitalHVO and senior medical scientists inthe HBB are important in raising theprofile of safe transfusion practice.• HBB should consider their practiceswith regards to technology, staffing,training and competencies in view ofrecent recommendations of UKCollaborative (Chaffe et al, 2009).34In terms of the HBB, a recent survey of thefunctionality of blood bank IT systems in the UKindicated that many of the systems wereinstalled before 2000 and have only limitedability to support and improve transfusionpractice (Murphy and Little, 2008). Similarly,restricted LIS are still in place in the vast majorityof hospitals in Ireland. The recommendations arethat HBB systems should interface with hospitalinformation systems and have the potential tointerface with bedside transfusion bloodsampling and administration systems, and thatthey should also interface with blood centresystems to facilitate imposed blood stockmanagement are fully applicable in Ireland. Thisrecommendation has received further support inthe recent report from the UK TransfusionLaboratory Collaborative (Chaffe et al, 2009).

Paediatric Adverse Events2008In 2008, the NHO received 24 reports relating toIBCT/SAE in paediatric patients

Blood to wrong patient (n=1)IBCT/SAE Case History 14In this case, two neonate twins requiringtransfusion were cared for in ICU. Nursingcare in the unit was delivered on a one toone basis. The nurse caring for twin Acollected the unit of red cells issued to twinB. This unit was transfused. The error wasdiscovered by a medical scientist in the HBBwhen the nurse caring for twin B came tocollect a unit for her patient. A review ofcontributing causes to this error identifiedthat both patient identifiers were extremelysimilar. Both twins were of same gender, hadidentical names (as neither baby had a firstname at the time) and dates of birth, andtheir medical record numbers were onlydifferentiated by one digit.Error cause: The root causes of this eventwere identified as human failures as follows;failure to follow policies – checking wasremote from the patient, failure to verify thecorrect details against the patient, andhuman slip - nurses at collection and atchecking missed the one digit difference inthe medical record number.Transfusion of antigen incompatible red cells(n=1)IBCT/SAE Case History 15An adolescent patient with an anti-Jk aantibody was transfused Jk a positive redcells. While this patient had previously beentransfused and had a Jk a antibody historynoted in the HBB, a flag identifying thishistory had been removed from the LIS.Further investigation failed to identify whenor how this had occurred.Unnecessary transfusion (n=1)This child received an unnecessary red celltransfusion, based on an incorrecthaematology result.IBCT/SAE Case History 16The event occurred when an incorrect Hbresult was released from the haematologylaboratory and resulted in a child receivingan unnecessary red cell transfusion. The Hbwas read on the laboratory analyser whichwas not in fact working properly, and themandatory repeat which was undertaken tocheck the aberrant result was performed onthe same analyser as the second analyserwas out of service.Incorrect Component Transfused (n=5)There were four cases of transfusion of incorrectred cells and one of platelets involving infants.Unnecessary Donor ExposureIn three cases involving red cells, infants receivedred cells, when there were assigned paedipacksin the HBB. This resulted in unnecessary donorexposure. In all cases, the crossmatched unitswere available for intraoperative use, andremained available in a satellite fridge after thepatient was returned to the clinical area. In two ofthese cases, first site of error wasprescription/request.Two system failures were identified ascontributing to one adverse event, a lack oftraining for doctors on prescribing red cells, andcustom and practice, where red cells available ina satellite fridge for theatre were used for top-uptransfusions to infants. In the final case the site offirst error was the HBB.In the fourth case, a unit of red cells wascrossmatched for an infant. In the interim period,the baby’s surname changed and when the redcells were issued to the clinical area, the nurselooking after the patient “updated” thecompatibility label with the baby’s name, insteadof returning the unit to the HBB.In the final case, an infant received pooledinstead of apheresis platelets. This adverse eventoccurred out of hours and the medical scientistinvolved did not normally work in the HBB.A further case of unnecessary donor exposurewas reported as an over-transfusion (IBCT/SAECase History 17).Over Transfusion (n=3)These events occurred at the administrationstage of the work process. A further episode of36

over-transfusion was collected as an unclassifiedSAR, (SAR Table 3: Symptoms associated withunclassified reactions page 49 (Case 4)).IBCT/SAE Case History 17A critically ill three month old infant in anintensive care unit should have received 90mls of red cells, but was instead transfused200mls. This occurred when a nurseincorrectly calculated the transfusion rate onthe infusion pump, thereby deliveringincreased volume to the infant. While thispatient suffered no apparent sequelae, thisover-transfusion would not have occurredhad he received a readily available paedipackin the laboratory. While it was not clearwhy this occurred, it is noted the patient hadbeen in theatre.Transfusion of incorrectly labelled units (n=3)There were three reports in which paediatricpatients received incorrectly labelledcomponents. These errors were not picked up atthe bedside.In one case, a medical scientist selected theincorrect donor number for label and printedthe label. While the hospital compatability labelcontained the correct patient demographicinformation, the unit number was incorrect.In the second case, a doctor working in A&Eincorrectly labelled a sample tube with theincorrect gender of the patient. The patient wasunconscious. This was an emergency situationwith multiple trauma and casualties.The third case is separately described.IBCT/SAE Case History 18A 14 year old child was transfused with anunlabelled cell salvaged unit of blood. Allcell salvaged units were routinely transfusedin theatre, but in this case the unit wasreturned to the clinical area. A nurse caringfor the patient transfused the unit. Thereporting establishment deemed the site offirst error to be administration, as the nurseshould not have transfused this unlabelledunit. This non mandatory IBCT wasdiscovered by the HVO.Error cause: There were clear systemfailures underpinning this event. Firstly therewas only one experienced technician tomanage intra-operative cell salvage in thehospital. In this case, he was unavailable andan inexperienced colleague managed thiscase. There was a failure of management toprioritise safety and invest in a full timedeputy for the technician. Secondly, therewere no policies in place to manage the cellsalvage process in the hospital. Furthermorethe patient was hurriedly returned fromtheatre and there was no supportingdocumentation or communication of theintra-operative cell salvage process.Failure to transfuse irradiated components(n=2)Both events involved red cells, and site of errorwas at prescription/request in one case and atthe HBB in the second case.Transfusion of incorrectly stored red cells(n=1)This event involved transfusion of red cells oversix hours following initial removal from storage.Other adverse events (n=5)The following events were included:• Use of a non-filtered administration set atadministration n=3.• Patient in Sickle Cell Crisis transfused withABO and RhD compatible red cells, but unitswere not phenotyped due to error in theHBB n=1.• Transfusion of red cells where pack wasperforated n=1.Specific paediatric practice issues highlightedin reports of 2008In 2008, there were two cases where neonatalpatients received blood components ofincorrect ABO group, thereby not meeting theneeds of these patients.Both laboratory and clinical practitionersworking in specialist neonatal and paediatricorganisations must be aware of the specialistrequirements of their patients.Patient identification is central to safetransfusion practice. There are certainidentification issues inherent in paediatricpractice, which make it more challenging e.g.the case of the twin who received the wrong37

lood, or where patients’ names are changed.Paediatric hospitals must be aware of theseissues ensure that patients are correctlyidentified. Staff caring for patients must bereminded that “short-cuts” may result in unsafepractice.In 2007 and again in 2008, the NHO reportedcases where neonatal patients are unnecessarilyexposed to further donors. In all cases, therewere assigned paedipacks already available forpatients. These errors tend to occur when bloodis available in both the HBB and in a satellitefridge for patients, for use during surgery, whichis not used and not returned to the HBB.Hospitals should examine their work process tominimise these occurrences. In one reportingestablishment, training and clinical updateshighlighting this issue were increased to theclinical areas involved. This has resulted in analmost complete reduction in these adverseevents in 2009.Again in 2008, the NHO noted reports of overtransfusionin paediatric patients. While therewere no associated reports of TACO, neonataland paediatric patients are at risk of such anevent. There was one report of an SAR in 2008resulting from over-transfusion. All errorsoccurred at administration, where incorrectcalculations resulted in these errors. In one case,an ambiguously documented prescriptioncontributed to the error.Key Points• Almost 22% of all reports of IBCT/SAErelated to paediatric patients.• Up to 12% of all IBCT/SAE reportsinvolved patients aged less than oneyear.• Analysis of IBCT/SAE in terms ofpotential to cause harm to patientsshowed that 63% of reportedIBCT/SAE had high potential to causeharm in paediatric patients, comparedwith 39% of reports in the adultpopulation, thereby highlighting risk topaediatric patients receivingtransfusions.• Paediatric patients are long termsurvivors of transfusion therapy. It isimportant that practitioners working inpaediatric centres continuously seek tominimise donor exposure to thesepatients.• Medical staff should ensure that prescribingrates and volumes of transfusions for thesepatients are unambiguous.• Nursing staff should carry out independentcross checks of rate and volume prior totransfusion.38

IBCT Involving FactorConcentrates/Blood Products2008The NHO collects incidents involving errorsinvolving factor concentrates and blood productsas they relate to transfusion practice. Incidentsinvolving anti-D are described in a separatechapter. Adverse reactions to factor concentratesand blood derived medicinal products fall underpharmaceutical legislation and as such aredirectly reportable to the Pharmacovigilancesection of the IMB and therefore are notdescribed in this report.Serious adverse events involving factorconcentrates are not reportable under the EUblood Directive. There were seven IBCT involvingfactor concentrates/blood products, one of whichinvolved a paediatric patient. There were nosequelae in any of the reported incidents.FindingsWrong Product (n=4)Four cases (57%) involved the wrong productbeing administered.In the first case a young adult with VonWillebrand’s disease presented at the emergencydepartment (ED) following a head injury. Thedoctor on duty administered Vasopressin(DDAVP) used in many patients with VonWillebrand’s disease to raise factor VIII levels.However, this particular patient does not respondto this product and should have received humanderived factor VIII. The error was discovered onreview of the case by the consultanthaematologist who prescribed the correctproduct. Analysis of the incident identified that acontributory factor was failure of the prescribingdoctor to contact the consultant haematologyteam for advice. In addition, the incidentoccurred out of hours when the patient’s medicalrecord was unavailable. An electronic record wasavailable but the prescribing doctor failed toaccess it. There were no sequelae as a result ofthis incident.The second case involved an elderly patient onwarfarin therapy (INR=7.9) for an underlyingcardiac condition who presented with dyspnoeaand wheeze following a collapse at home. Thepatient received recombinant Factor VIIa whenthe correct product should have been vitamin K(as stipulated in hospital guidelines). This wasgiven the next day. The error was discoveredretrospectively by the HVO. While there were nosequelae as a result of this error, the patient wasexposed to recombinant Factor VIIa which shouldnot be used outside the recognised indicationsbecause of high risk of complicationsThe remaining two cases were due toinappropriate administration of PCC. One caseinvolved an elderly patient with multiplepathologies who was given PCC eight daysfollowing a surgical procedure. The patient wasnot actively bleeding, but her blood plateletcount and coagulation screen were abnormal(Platelets=32: Activated Partial ThromboplastinTime (APTT) =150: INR =5.0). The product wasprescribed following advice from thehaematology registrar but the consultanthaematologist, on review, decided this wasincorrect.The second case was attributable to failure toadhere to the hospital guidelines in place at thetime. The prescribing doctor ordered PCC forpost-operative bleeding (INR of 2.3) for a patientnot on Warfarin. As the hospital policy has sincebeen amended to allow the use of PCC inmassive haemorrhage if the patient continues tobleed after administration of SD Plasma,retrospectively this cannot be considered as anIBCT, but was collected as such at the time.Wrong Dose (n=3)Three cases (43%) errors involved the wrong doseof product being given. In one case theconsultant haematologist gave a verbal order fora patient with a severe factor VIII deficiency to begiven 3,000 international units (IU) of factor VIII.Prior to administration the product and dosagewas checked by two nurses. Duringadministration, however, the nurse realised she39

had made an error and stopped the infusionbut by then, 3,500 IU had been given.Contributory factors to this incident were thatthere was no written prescription for theproduct and the second nurse checking thedosage was unfamiliar with the product. Therewere no complications as a result of thisincident.In the remaining two cases the patientsreceived a smaller dose than had beenprescribed. In one case, a patient with a severefactor VIII deficiency had a gastro intestinal (GI)bleed. The doctor calculated that the patientrequired 4,000 IU of factor VIII. However, whenremoving the vials she took the wrong size vialfrom the fridge and administered only 3,500 IU.The error was discovered retrospectively by theHVO. Investigations identified that although thedoctor checked the product with a nurse priorto administration, the nurse was unfamiliar withthe product, and secondly, this event occurredat the end of a shift. There were no sequelae asa result of this incident. The second case isdescribed in IBCT/SAE case history 19.IBCT/SAE Case History 19This case involved an infant with factor VIIIdeficiency. The infant had undergone asurgical procedure and was prescribed acontinuous infusion of factor VIII (Advate) 4IU per kg/per hour over a five day period. Inerror, the infusion was set at 2 IU per kg/perhour. This was not discovered until 48 hourslater when a HVO was training clinicalpractitioners and noticed the infusion pumphad been set at the incorrect rate. Therewere no sequelae as a result of this undertreatment. A contributory factor was thatthe prescribed rate of infusion for thisproduct was different to that normally usedin this clinical situation. In addition, theincident occurred over a weekend whenstaff familiar with this product and protocolswere off-duty.the 2007 NHO Annual Report. The risk oferrors when administering factorconcentrates is a constant hazard,particularly if staff are unfamiliar with theavailable products. To reduce this risk thereneeds to be systems to ensure the correctproduct is administered to the correctpatient. The National Centre for HereditaryCoagulation Disorders (NCHCD) hasproduced a standard protocol for staffadministering factor concentrates. This isavailable from the NCHCD, located in St.James’s Hospital, Dublin 8.• Where possible, staff familiar withcoagulation and blood products should beinvolved in their checking andadministration. If clinical staff are unfamiliarin treating patients with coagulationdisorders, they must seek advice from acoagulation specialist.• Verbal orders for coagulation factors andblood products are to be avoided. Insteadthere must be a written prescriptionreducing the risk of error.• If a hard copy of the patient’s record isunavailable out of hours, then staff shouldbe trained and aware of where to find theelectronic record. All staff involved in theprescription, issue and administration offactor concentrates and blood productsshould receive appropriate training.Recommendations• Several cases described in this reportoccurred due to lack of knowledge,unfamiliarity with the different products anda failure of clinical staff to adhere to policiesand guideline. This was also highlighted in40

Adverse Events associated withAnti-D Immunoglobulin2008Incidents involving errors or omissions relating toanti-D Immunoglobulin (Ig) should be reported toNHO as non-mandatory IBCT as they relate totransfusion practice. They should also bereported internally within the hospital riskmanagement procedures.Adverse reactions related to the administration ofanti-D Ig are reportable directly to the IMB underthe Pharmocovigilance Scheme and if received bythe NHO, are forwarded to the IMB. These aretherefore not covered in this report.Serious adverse events related to anti-D Ig arenot reportable under the EU Blood Directive.Findings:There were 29 reports of serious adverse eventsassociated with anti-D Ig administration. Thisrepresents an increase of 45 % compared to2007. Twenty-eight cases were categorised aslevel 1 or high risk that is having real potential forsensitisation or harm. The remaining case wascategorised as level 2 or moderate risk meaning itis unlikely to cause permanent harm to themother or foetus immediately in the future as thepatient did not fit the criteria for anti-D Ig.The majority of errors were due to omission ordelay in administering anti-D Ig, and most errorsoccurred in the clinical area as shown in Anti-DFigure 1. In some cases the error involved morethan one department or person.Anti-D Figure 1: Where error occurred (n=29)3 (10%)2 (7%)2 (7%) 2 (7%)ClinicalClericalLaboratoryPatient-relatedUnclear20 (69%)Omissions related to failure to give anti-D Ig inantenatal patients (n=12)In 12 cases, anti-D Ig should have been given butwas omitted. All cases were associated withhuman error and in two cases system failureswere also identified. The reasons for the errorsare shown in Anti-D Table 1. In two cases morethan one error cause was identified.Anti-D Table 1: Breakdown error causes inanti-D Ig IBCT (n=14)Human Failure n System Failure nKnowledge 5 Culture 1Co-ordination/ 1 Policies/Procedures 1CommunicationFailure to adhere 3to policies andproceduresUnclassifiable 1Patient-related 1Slip error 1Omissions related to failure to give anti-D Ig inantenatal patients (n=12)All 12 cases involved Rh D negative antenatalpatients who presented with bleeding per vagina(PV) or following trauma. Ten of these casesoccurred in general hospital emergencydepartments (ED) and it is known that in at leasttwo of these cases the patient has subsequentlybecome sensitised.In one case an Rh D negative patient attended forher first antenatal visit at 15 weeks gestation andreported that she had PV bleeding a few daysearlier. She was grouped as Rh D negative and noantibodies were detected. Anti-D Ig was notadministered but it was unclear if the omissionwas due to failure to follow-up on the result ordue to lack of knowledge of the doctor at theantenatal clinic. Anti-D was detectable in herserum five months later. The patient subsequentlydelivered an Rh D positive baby but anti-D Ig was41

not given as the patient had already beensensitised.In a further case, both system and patient-relatedfactors were involved when there was a delay inthe doctor attending to administer anti-D Ig. Thepatient went home and refused to return fortreatment. In the remaining cases either thepatient’s blood group was not checked or thejunior doctors on duty felt anti-D Ig was notindicated despite the availability of hospitalguidelines.Anti-D Case History 1In another case an Rh D negative mother of34 weeks gestation attended an ED with PVbleeding. An emergency sample was manuallygrouped as Rh D positive. However, the nextday during checking of the results of ‘on call’work on the blood grouping analyzer, thepatient was found to be Rh D negative. At thetime of the incident there was no link betweenthe manual and automated blood groupresult, this has since been corrected.Additional contributory factors were that thepatient had two records with two differenthospital numbers. In one chart it was notedthe patient had attended the hospitalcomplaining of a urinary tract infection with nomention of bleeding. In addition, the resultsof the emergency blood sample were filed inthe wrong chart. The anti-D Ig omission wasonly discovered by chance when the patientwas in labour, and the midwife happened tomention the error to the HaemovigilanceOfficer.Delay in administering anti-D (n=13)In a further 13 cases, anti-D Ig administration wasdelayed beyond the recommended limit of 72hours following a sensitising event. In one casethere was no documentation to indicate why theerror occurred. Human error was involved in all ofthe remaining 12 cases, and in two cases systemerrors were also identified. The findings of rootcause analysis of these incidents are shown inTable 2. In six incidents more than one error causewas identified.Anti-D Table 2: Delay of anti-D Ig: Error cause(n=19)Human Failure n System Failure nFailure to adhere 5 Culture 1to policies andproceduresKnowledge 1 Design 1Verification 1 Policies/Procedures 1Co-ordination/Communication 8Carrying out taskincorrectly 1Anti-D Case History 2In this case the administration of anti-D Ig wasdelayed due to a difficulty in interpreting thecord bloods. Following delivery of a baby toan Rh D negative mother cord blood sampleswere sent to the laboratory at a weekend. Thehospital blood bank scientist interpreted thecord blood sample as Rh D negative but wasnot completely satisfied with the result. Arepeat sample showed a similar result,therefore cord blood samples and a samplefrom the baby were sent to the referencecentre for further analysis. The referencecentre found the cord sample to be Rh Dpositive (weak D type) but the delay resultedin the anti-D Ig being administered more than72 hours post-delivery.Anti-D Case History 3In another case an Rh D negative mother wasgiven anti-D Ig one day prior to delivery for anante partum haemorrhage. The next day themother gave birth to an Rh D positive baby.However, this was not seen as a furthersensitising event requiring anti-D Ig.Consequently anti-D Ig was not given untilmore than 72 hours following delivery.Unnecessary Treatment with anti-D Ig (n=4)There were four cases of unnecessary anti-D Igadministration to Rh D negative mothers. All ofthese cases were found to be due to human error,more specifically lack of knowledge, and in twocases a failure of communication contributed tothe error.42

Anti-D Case History 4 & 5In two cases anti-D Ig was given to previouslysensitised patients. In the first case a patienthad suffered a sensitising event several yearspreviously and had developed antibodies.There was a record of this in the hospitallaboratory records. At 24 weeks gestation inthis pregnancy, following a fall, she attendedthe ED and was given anti-D Ig withoutchecking her blood group or antibody status.The incident occurred at a weekend when it ispractice at this hospital for clinical staff tocollect anti-D Ig from the laboratory, and thepatient’s blood group and antibody status werenot checked prior to administering the anti-DIg.In the second case an Rh D negative motherwith known anti-D and anti-E antibodies wasgiven anti-D Ig post-delivery of an Rh Dpositive baby. This indicated a lack ofknowledge on the part of the clinical staff. Acontributory factor was that the medicalscientist did not question the issue of the anti-D Ig. This practice is now under review.Anti-D Case History 6In another case an Rh D negative patient wasfound to have developed anti-D in her serum at29 weeks gestation. The anti-D was quantitatedon three further occasions by the referencelaboratory and referral for fetal medicineassessment had been advised. At delivery afurther maternal sample was taken and referredfor quantitation. The baby was DAT positiveand grouped as Rh D positive. However anti-DIg was issued by a medical scientist who hadjust recently completed training. The error wasdiscovered by another medical scientist whenupdating the patient’s record.Anti-D Case History 7The final incident involved a patient whounderwent a caesarean section and subsequenthysterectomy for post-delivery complications.Anti-D Ig was issued from the blood transfusionlaboratory who had not been informed that thepatient had had a hysterectomy. Clinical staffwere unsure whether to administer the anti-DIg. The hospital did have anti-D Ig guidelinesbut they did not indicate what action should betaken for Rh D negative patients posthysterectomy. After a clinical review it wasdecided to administer the anti-D Ig. Thehospital guidelines have since been amendedto include guidance on indications for anti-D Igin Rh D negative women post- hysterectomyand tubal ligation.Follow up actionIn a number of cases no changes to practice werereported. For those hospitals that did introducechanges several amended their trainingprogrammes for clinical staff on the indications forthe administration of anti-D Ig. Other changesincluded:• Better communication / informing all staff anddepartments involved in the early discharge ofpost-natal patients to ensure follow-up of Rh Dmothers requiring anti-D Ig after discharge.• Rechecking of manual and automatedgrouping.• At one site where it is routine practice fordoctors to administer anti-D Ig it is proposedthat midwives give anti-D Ig to avoid delays inpatients receiving anti-D Ig.• Where ante-natal patients present with ahistory of trauma or bleeding in pregnancyhospital policies have been changed toindicate that they should not be dischargedbefore their Rh D status has been establishedand the appropriate treatment given.Key Points• A number of hospitals still do not haveguidelines for the administration of anti-D Ig. Each hospital should have clearwritten protocols indicating when andhow to administer anti-Dimmunoglobulin.• Lack of communication or co-ordinationwithin or between departments orindividuals were notable factors in thedelays and omissions of anti-D Igprophylaxis cases. Similar findings werereported in 2007.43

• Two cases of anti-D Ig sensitisationreported this year were attributable to afailure to administer anti-D Ig during apregnancy. Several incidentsdemonstrated a lack of understanding orfailure to adhere to guidelines foradministration of anti-D Ig. Despite thisbeing highlighted in previous NHOreports there still appears to be a deficitof knowledge among clinical staff ofwhen to administer additional anti-D Igfor sensitising events following previousantenatal administration. There is a needfor specific education in those instanceswhen Routine Ante-natal Anti-DProphylaxis (RAADP) is introduced(NICE, 2002; BCSH, 2006a).• Two cases relating to unnecessary anti-DIg reflect a lack of knowledge andprotocols for the management ofsensitised pregnancies and raise concernover possible failures to monitorpregnancies at risk of HDN.Recommendations• There should be clinical follow-up andretesting in 6 months of patients in whom anti-D Ig administration has been delayed oromitted. Any sensitisations arising should bereported to the NHO as well as internally.Similarly a review of the previous obstetricalhistory should be performed to determine ifanti-D Ig was administered in accordance withprophylaxis guidelines in all new patientspresenting with immune anti-D in their serum.• The requirement for anti-D Ig prophylaxismust be included in the discharge checklistprocedure for Rh D negative women,particularly in the case of early post-nataldischarges.negative card (BCSH, 2006b) with listedindications for prophylaxis to the patient.• If the HBB Rh D typing results are inconclusiveor discrepancies are identified then thereshould be a protocol indicating the procedureto follow. As noted in previous NHO reports(2004) such samples should be referred to areference laboratory but the hospital musthave systems in place to ensure that thosepatients requiring anti-D Ig receive it and thatit is not omitted or delayed.• Clear procedures for communication of therequirement for anti-D Ig prophylaxis shouldbe implemented.• Hospital guidelines need to clearly state thatanti-D prophylaxis should be administered inresponse to sensitising events regardless ofrecent or planned administration ofprophylactic anti-D.• Anti-D Ig should only be issued byexperienced trained laboratory staff who areable to interpret the results.• Training of laboratory staff on when anti-D Igshould be administered is also required. An e-learning module for laboratory and clinicalstaff on anti-D Ig and antenatal testing is inpreparation and will be added to thee-learning programme in 2010(http://www.learnbloodtransfusion.org.uk) It isrecommended that these modules are madeavailable for training of clinical and scientificstaff involved in antenatal testing and anti-D Igadministration and that they perform thecompetency assessments within the modules.44• The patient should be informed of her Rh Dnegative status and situations requiring anti-Dprophylaxis.• There should be clear identification of thepatient’s Rh D negative status in both hospitaland shared care files.• Hospitals should consider using a Rh D

Serious Adverse Reactions (SAR) 2008There were 196 SAR reports during the reportingyear 2008. After review, 143 SARreports were accepted by the NHO and theremaining 53 did not progress (DNP) as they didnot meet reporting criteria. As in previous years,there were no reports of Transfusion AssociatedGraft versus Host Disease (TAvGHD), PostTransfusion Purpura (PTP) or adverse donorreactions related to predeposit autologoustransfusion (PAD). There was one report receivedas TRALI.Acute Transfusion Reaction (ATR)ATR are defined as those occurring within 24hours of transfusion. Acute reactions include:• Acute Haemolytic Transfusion Reactions• Febrile Non-Haemolytic Transfusion Reactions• Acute Allergic and Anaphylactic TransfusionReactions• Hypotensive Reactions• Unclassified ReactionsDuring the reporting year 2008, 89 reports of ATRwere reported. The breakdown is given in thetable below.SAR Table 1 (n=89)Serious Adverse ReactionTotal numberprogressedAHTR 2FNHTR 38Acute Allergic and AnaphylacticTransfusion Reactions 41Unclassified 6Hypotensive 2Total 8945

Acute Haemolytic Transfusion Reactions(AHTR) (n=2)2008Definition: Acute Haemolytic Transfusion Reaction(AHTR) is defined as a reaction occurring within 24hours of a transfusion where clinical and/orlaboratory features of haemolysis are present(International Society of Blood Transfusion (ISBT)Working Party, Capetown, 2006). Acute haemolysismay be caused by ABO incompatibility, otherantigen incompatibility e.g. RhD, Kell or to nonimmunologicalfactors such ashypertonic/hypotonic solutions or medicinalproducts mixed with the blood component.EU Notification CategoryThese AHTR are reportable to the EU as• Immunological haemolysis due to ABOincompatibility• Immunological haemolysis due to otheralloantibody - Acute• Non–immunological haemolysis.FindingsTwo AHTR reactions were reported in 2008. Bothreactions were immunological haemolysis due toother allo-antibody. In the first case the patientwith a history of anti-Fy a antibody was transfusedwith a Fy a positive unit due to an error in the HBB(SAR Case History 1).Immunological haemolysis due to other alloantibodyacuteSAR Case History 1This patient with a history of anti-Fy a antibodiesrequired a transfusion of two units of red cellsout of hours. Within one hour of commencingthe transfusion the patient had a reaction. Oninvestigation it was discovered that the patienthad been transfused in error with a Fy a positiveunit due to an error in the antigen typing.Error Cause: The error happened on call overthe weekend. The medical scientist who rotatedthrough the HBB on a regular basis carried outthe antigen typing incorrectly due to humanerror. Also due to a system failure there wasinsufficient documentation of the technicalresults during the procedure to enable checkingof the scientist technique and results.Changes to practice: As a result of the error,there was retraining of staff, amendments to thepolicies and procedures and a new stepwiserecording system for all technical workimplemented.SAR Case History 2In the second case the patient had both anacute and delayed reaction. This case involvedan elderly male patient who had requiredtransfusion post operatively and required furthertransfusion support. A pretransfusion samplewas sent and was found to be DAT positive (1+).(IgG(+w) and C3d(1+)). The patient had an anti-D and anti-Fy a identified but a non specificantibody reacting in IAT and Enzyme panels wasalso detected.The least incompatible red cells RhD negativeFy a negative were issued for this patient andtransfused the following day. Just over threehours into the transfusion the patient developedsymptoms of hypertension, pyrexia and chills.On investigation of this reaction the patient hada raised bilirubin level, DAT was negative andantibody specificity on the post transfusionsample remained undetermined.Ten days later the patient was readmitted forinvestigations and it was noted that there hadbeen a fall in Hb associated with a transientelevation of bilirubin. The DAT was still negativeand antibody investigation was inconclusive.Samples were sent to a Reference Laboratoryfor further investigation an anti-K, anti-Jk a ,anti-Fy a and anti-S were identified.Clinical OutcomeBoth patients made a complete recovery.Key Points• Although it may be necessary to issueleast incompatible blood, in anemergency, samples should be sent to areference laboratory for confirmation.46

Febrile Non Haemolytic TransfusionReactions (n=38)2008Definition: FNHTR is defined as a rise intemperature of >1.5 o C above the patient’s(pretransfusion) baseline value, together withrigors or chills, rarely nausea, vomiting ordyspnoea occurring during or within four hoursfollowing transfusion without any other cause,such as haemolysis, bacterial contamination ordue to the patient’s primary diagnosis. Althoughtraditionally counted as unpleasant, but notserious, as patients usually recover quickly,FNHTR can be upsetting for the patient and mayrecur on further transfusions.EU Notification CategoryFNHTR is incorporated in the following EUcategory: Other – FNHTRFindingsThere were 38 reports which fulfilled the criteriafor FNHTR, an increase of eight reports on 2007.Thirty seven of the patients experiencing aFNHTR were adults, the majority of these wereelderly and one further reaction occurred in achild. Thirty five reactions were associated withred cells and one each with apheresis platelets,pooled platelets and granulocytes. The majorconcern in evaluating these reactions is toexclude bacterial contamination of the unit orhaemolysis due to incompatibility (Heddle &Kelton, 2001).In addition to the 38 FNHTR reports received,four further cases which presented with febrilereactions were captured as STTI (See cases 1,2 3,and 4 SAR Table 10 STTI Bacterial).Clinical OutcomeA clinical outcome was given for all but one case.The majority of patients (32) made a completerecovery, with the timeframe of recovery given in23 cases. Of these 23 cases, the majority ofpatients recovered within five hours, one patientrecovered within seven hours and the final patienttook eight hours to recover. Four patients sufferedminor sequelae as a result of the reaction; two ofthese patients required overnight admission, thethird patient took 24 hours to recover and in thefourth case, the recovery time was not given. Onepatient died unrelated to the transfusion.Analysis of the cases reported indicates that themajority of the cases were investigated forbacterial contamination. Thirty one (81.5%)patients had blood cultures taken. In 29 (76%)cases the pack was sent for culturing and in 26(68%) cases both the patient and pack werecultured. However, only 15 (39%) cases wereinvestigated for haemolysis and only 13 (34%)cases in total were investigated for bothhaemolysis and bacterial contamination.47

Acute Allergic and AnaphylacticTransfusion Reactions (n=41)200848Clinical Signs & Symptoms and LaboratoryDefinitionsAllergic and anaphylactic transfusion reactionsspan a range of symptoms of varying severity.• The symptoms encompass mild allergic-typereactions such as urticaria/pruritis associatedwith or without gastrointestinal discomfort, tomajor reactions with stridor, wheeze,angioedema, bronchospasm and hypotensionoccurring during or within four hours oftransfusion (ISBT, 2006).• An anaphylactic reaction or anaphylaxis ischaracterized by severe hypotension andcollapse which may be accompanied bylaryngeal oedema and respiratory obstruction(Povosky, 2001).• Tryptase levels if available prior to thetransfusion and within 2-3 hours of thereaction taking place, may help to confirmdiagnosis.• Allergic type reactions apart from pruritis, mildrashes or urticaria associated with transfusionshould be submitted to the NHO.• A small number of patients with severe IgAdeficiency develop antibodies to IgA. Someof these patients may have severe anaphylaxisif exposed to IgA through transfusion(McClelland, 2001).EU Notification CategoryAA reactions are reportable to the EU asAnaphylaxis/Hypersensitivity.FindingsThere were 41 reports which fulfilled the criteriafor this reaction. Twenty seven (27) reactionsoccurred in adults and 14 occurred inpaediatric/adolescent patients. Eleven (27%)reactions were associated with red cells, 15 (37%)reactions were associated with apheresisplatelets, 13 (32%) reactions to pooled platelets(three reactions to pooled platelets in plasma, tenreactions to pooled platelets in PAS) and one (2%)reaction each to cryoprecipitate and SD plasmarespectively. Similar to last year, no cases of AAdue to IgA deficiency with antibodies werereported, but IgA levels were only reported in 25cases.SAR Table 2: Acute Allergic and AnaphylacticReaction per type of platelet componentissued 2008 (n=28)Type of platelet issued NumberissuedIncident ofreactionper unitissuedPlatelets pooled in PAS 9,848 1 per 984unitsissuedLeucodepleted Plateletspooled in plasma 1,009 1 per 336unitsissuedApheresis Platelets 13,754 1 per 917unitsissuedAs in previous years, the majority of thesereactions were associated with platelets with thehighest reaction rates associated with plateletspooled in plasma and with much lower ratesassociated with pooled PAS platelets andapheresis platelets.Clinical OutcomeThe clinical outcome was given in all but threecases. The majority of patients (35) made acomplete recovery, with a number requiring minortreatment. The final three patients had minorsequelae, with two requiring overnight admissionas a result of the reaction. The final patientrecovered following symptomatic treatment.

Unclassified Reactions(n=6) 2008Definition: Unclassified SAR is the occurrence ofnew adverse symptoms/signs with no risk factorother than the transfusion and which on its own doesnot allow the reaction to be classified within thedefined categories of SAR.FindingsA total of 16 reactions were originally submitted asunclassified reactions. Following review, six caseswere recategorised (two FNHTR, two AA, one TADand one Hypotensive reaction), four cases did notprogress and six reactions were accepted asunclassified. Four of the patients were adults andtwo were paediatric patients. Four reactions wereassociated with red cells and two reactions wereassociated with apheresis platelets.The table below shows the symptoms each patientexperiencedSAR Table 3 – Symptoms associated with Unclassified Reactions (n=6)Clinical OutcomeFour patients made a complete recovery followingminor treatment. The fifth patient (Case 4) inpaediatric section who was a young child and wasover transfused required a venesection, but fullyrecovered following this procedure and the sixthpatient (Case 2) died unrelated to the transfusion.Recommendations for management of ATR• Whenever possible, as a minimum, bloodcultures and investigations for haemolysis shouldbe taken on patients suffering a FNHTR toexclude red cell incompatibility or bacterialcontamination.• Reaction Alerts in patient charts and/or on thehospital patient admittance system and ITsystem can be valuable in those patients with aprevious AA/FNHTR reaction to ensureappropriate component selection and premedication prior to future transfusions.Case Age Component Temperature Hypertension Falling O2 Substernal Restlessness Other DetailsNo Rise Saturation discomfort /anxiety1 Child RCC Yes Yes Yes Leg pain, pins and(5-11 needles andyears)discomfort2 Elderly RCC Yes Yes Consolidation(70+) present on chestXray prior totransfusion3 Adult RCC Flushing along(51-70 infusion siteyears)4 Neonate RCC Yes Plethoric/bronze(

Hypotensive TransfusionReaction (n=2)2008Definition: Characterized by a drop in systolicand/or diastolic blood pressure of >30mmHgoccurring during or within one hour ofcompleting transfusion with no othersymptoms.FindingsThis is the first year the NHO has collected thistype of reaction. Two cases that fulfilled thereporting criteria were accepted by the NHO.Both cases involved neonates with underlyingcomplex cardiac conditions. The implicatedcomponent in both cases was cryoprecipitate.SAR Case History 3In the first case a neonate was bleedingwhile undergoing surgery. Within twominutes of receiving the component, thebaby developed severe hypotension andresponded in a similar manner when given asecond dose. When the infusion wasstopped, the blood pressure returned tonormal. It was felt that the reaction wasclearly related to the cryoprecipitate andwas due to citrate toxicitySAR Case History 4In the second case the patient had recentlyreturned from theatre to the intensive careunit (ICU) following complex surgery andwas bleeding. The cryoprecipitate wasbeing administered when the patient wentinto cardiac arrest. Again, the reactionwhich was probably related to thecryoprecipitate was considered due tocitrate toxicity and a decision was taken notto administer cryoprecipitate to this patientagain.Clinical OutcomeIn the first case the patient recovered when thetreatment was discontinued. In the second casethe patient had serious sequelae as the patientwent into cardiac arrest but subsequentlyrecovered.50

Delayed Haemolytic TransfusionReactions (n=4)2008Definition: DHTR are defined as evidence ofclinical or laboratory features of haemolysisoccurring more than 24 hours and up to 28 daysfollowing the transfusion of a blood component andassociated with serological evidence of antibodies.(ISBT, 2006)For the purpose of analysis, the NHO grades suchreactions by severity using the SHOT criteria.(SHOT, 1999) These are:Group 1:Group 2:Asymptomatic with ‘antibody only’detected, with or without a positiveDAT.Demonstrates evidence of haemolysismeasured by falling haemoglobinlevels and a positive DAT.SAR Table 4 Details of DHTR Reported (n=4)Group 3:Group 4:Evidence of a falling Haemoglobinlevel associated with jaundice, withor without a positive DAT level.Graded as for Group 3, but withassociated renal impairmentGroup 1 (no evidence of haemolysis) reactions arenot reportable to the NHO, it is however,recommended that a record of all Group 1 reactionsare maintained at hospital level.FindingsThere were four reactions which fulfilled reportingcriteria in this category in 2008. All the patientswere elderly. It was unknown in three of the cases ifthe patient had a previous transfusion history.Again, this year, unfortunately investigations forhaemolysis were incomplete. It would appear thatmany hospitals do not perform LDH levels routinelyas part of their general biochemistry screen.Case Previous Days postNo. Age Gender Imputability Findings Antibody Category Outcome Transfusion transfusionHistory1 Elderly Male Possible No LDH Anti M Group 3 Death Unknown 2-8 days(70+) available, unrelatedfall in Hb,to transfusionjaundice,positive DAT2 Elderly Male Likely/ Elevated LDH, Anti Kell Group 3 Death Unknown 6 days(70+) Probable DAT positive unrelatedpre/postto transfusiontransfusion,jaundice,no Hb levelat time ofreaction3 Elderly Male Likely/ Fall in Hb, Anti Jk a Group 3 Death Yes 7-14 days(70+) Probable no DAT, unrelatedelevated LDH,toelevatedtransfusionbilirubin4 Elderly Female Likely/ Elevated LDH, Anti Fy a , Group 2 Complete Unknown 8-18 days(70+) Probable fall in Hb Anti S Recoverypositive DAT.51

In Case 3, the patient required an emergencytransfusion of four units of uncrossmatched bloodfor haemorrhage. The patient had a previouslydetected Jk a antibody present. It was subsequentlydetermined that the first unit transfused was Anti-Jk a positive. Although the presence of Jk aantibodies was known, in view of the urgency of thesituation a correct clinical decision was made totransfuse the unit.Clinical OutcomeIn three cases the patients recovered from thereaction but died due to their underlying condition,the final patient made a complete recovery.Recommendations• It is likely that DHTR is under diagnosed. It isessential that any patient presenting withunexplained anaemia some days after atransfusion should be investigated forimmunological haemolysis (bilirubin, LDH, DAT,haptoglobulins and antibody screen) to excludeDHTR. In a number of the reports of DHTR in2008 the investigation was incomplete. Thesuccessful diagnosis also depends on accuratehistory taking and the eliciting of a history ofrecent transfusion.• The NHO has, in previous reports, suggestedthe development of a national patient antibodyregister for patients with red cell antibodies.• Patients who have had previous pregnancies ortransfusions are at risk of developing antibodies.Very often the HBB can be unaware of patients’history and the potential for antigenincompatible transfusions can be high.Development of a national antibody registercould address this risk by ensuring access topatients’ antibody history. It would also reducethe requirement for repeat laboratory testing.• This would only be feasible with theimplementation of a national Unique HospitalIdentifier (UHI) , a recommendation made byHIQA (2009) and supported by the NHO. A UHIwould facilitate improved and safer access topatients’ records on a national antibody registerthereby ensuring safer transfusion practice forpatients.52

Respiratory Complications ofTransfusion2008Transfusion AssociatedCirculatory OverloadNHO DefinitionTACO is characterised by the development of acutepulmonary oedema secondary to cardiac failure.Signs and symptoms can manifest during, or withinsome hours of transfusion and can include any or allof the following: dyspnoea, orthopnoea, cyanosis,tachycardia hypertension and pulmonary and/orpedal oedema. Chest auscultation reveals thepresence of rales (Popovsky, 2001).EU notification Category: Other TACOFindingsThere were 39 reports of TACO, accounting for 13%of reports of serious adverse reactions accepted bythe NHO. This is over 100% increase in the numberof reports when compared to 2007. There were 33cases associated with red cells, four with multiplecomponents and one with apheresis platelets. Afurther case associated with SD plasma was initiallythought to be a TRALI but on subsequentinvestigations was reclassified as TACO (SAR CaseHistory 7). Twenty seven (69%) were attributed aslikely/probable to the transfusion.Symptoms and underlying conditionsOnset of overload developed 15 minutes to 13hours after transfusion with a median onset of threehours. Falling oxygen saturation levels werereported in 20 cases and 28 (72%) of the patientshad complex medical problems. While the majorityof cases were associated with red cells, one casewas associated with apheresis platelets in a bonemarrow transplant patient with multiple comorbidities.The most commonly reported symptoms areoutlined in SAR Table 5.SAR Table 5 Most frequently occurringsymptoms in TACO (n=39)SymptomnDyspnoea 29Falling O2 Saturation 20Hypertension 15Tachycardia 14Underlying conditions were reported in 20 patientsand one patient was reported as having underlyingcardiac, respiratory and renal conditions. Therecovery time-frame for patients with multipleunderlying conditions was considerably greaterthan those who had only one or no underlyingcondition. Four patients took four days to 12 weeksto recover and three of these had a pre-existinghistory of cardiac, respiratory or renal conditions.Ten patients recovered within one to eight hourswith five of these having no or only one pre existingcondition.SAR Table 6 Underlying conditions of patientswho developed TACO (n=47)Underlying conditionNo of patientsCardiac 26Respiratory 9Renal 12The ISBT definition of TACO is more restrictive thanthat accepted by the NHO. The ISBT definition(2006) requires any four of the following fivesymptoms/signs occurring within six hours ofcompletion of transfusion:• Acute respiratory distress• Tachycardia• Increased blood pressure• Acute or worsening pulmonary oedema onfrontal chest radiograph• Evidence of positive fluid balanceThe number of TACO cases accepted by the NHOwould reduce greatly if the ISBT criteria wereadopted. This is evidenced by the fact that only one53

54case received by the NHO in 2008 met their strictcriteria. However this may be due to the fact thatonly 10 patients had a chest X-ray and 12 patientshad a completed fluid balance chart. In 27 casesthe fluid balance record was not complete and 17of these recorded input only.TACO in elderly patientsThe majority of reports of TACO (69%) occurred inthe elderly. The age and gender of the patientsimplicated in these reports are outlined in SARTable 7. The patient’s weight was not recorded in16 cases and in seven cases the weight was lessthan 60kg. Twenty patients were on regulardiuretics and of these ten received diuretics pretransfusion,four during transfusion and 16 posttransfusion.Twenty patients had received other components inthe 24 hours prior to transfusion. TACO occurred infour cases, three of whom had received red cells inthe previous 24 hours where only 50-100mls of theunit involved in the reaction had been transfused.Diuretics were administered pre and duringtransfusion to two of these patients, emphasisingthe need for close monitoring of elderly patients.SAR Table 7 Age/gender of patients implicatedin TACO reports 2008 (n=39)Adult Adult Adult Elderly(18-30 (31-50 (51-71 (70+)years) years) years)Male 0 1 4 13Female 1 4 3 13TACO in bleeding patientsThere were three cases where patients with severehaemorrhage requiring massive transfusiondeveloped TACO; two of these patients had nounderlying condition. The third patient had acuterenal failure which may have contributed to theTACO. While TACO in young previously healthyadults is unusual, there have been six similar casesreported to the NHO between 2000 and 2008. Fivecases were in females less than 30 years of age withmassive haemorrhage associated with obstetric andgynaecological complications and the sixth in a 41year old male. Two of the six cases were reported in2008 (SAR Case History 5 & 6 ).SAR Case History 5 (TACO)A 21 year old female patient with massive postpartum haemorrhage and Hb of 3.9g/dl received13 units of red cells, five platelet pools, 14 unitsof SD plasma, six pools of cryoprecipitateincluding colloid and crystalloid giving a total of13 litres of fluid over six hours. Her estimatedblood loss was 3,580mls and her urinary outputwas 2,800mls giving an overall positive balanceof 6,620mls. Frusemide was given and she wastransferred to the intensive care unit (ICU) inanother hospital.On arrival in ICU, the patient was hypertensiveand oedematous and her O2 saturations weredecreased. She had bilateral creps onauscultation of the chest and the chest X-rayshowed some shadowing throughout both lungsconsistent with a degree of pulmonary oedema.She was commenced on Continuous PositiveAirway Pressure (CPAP). During the next fewdays her urine output was poor, her creatininewas rising and she was in a positive fluid balanceof 2-4L. On day three, her chest X-ray stillshowed pulmonary oedema. Frusemide wasprescribed but not given as the patient startedto produce large volumes of urine and wasweaned from the ventilator. The retrospectivelytested pro Brain Type Natruiretic Peptide (proBNP) pre-transfusion was 103 pg/ml and posttransfusion was 356 (ratio 3.4) on day ofadmission to ICU but rose to 950 on the nextday.SAR Case History 6 (TACO)A 41 year old patient with severe epistaxisreceived nine units of red cells and four units ofSD plasma over a timeframe of five hours in thetheatre. The patient developed a pyrexia,hypertension, stridor, wheeze and chest X-raychanges. The fluid balance was incomplete buthe had an estimated intake of 10,000mls. Hemade a complete recovery after four hours withpulmonary oedema resolving withoutintervention and he did not require ICUadmission.

TACO with SD plasma – Still a problem!SAR Case History 7 (TACO): This femalepatient (54kg) required four units of SD plasmafor an INR of 1.5 which had not reduced withVitamin K pre liver biopsy. The transfusion wasprescribed over 3-4 hours. Due to poorintravenous IV access, there was a delay inadministration of the first two units and in orderto avoid wastage of thawed units all four unitswere administered in 1 to 1.5 hours instead of atthe prescribed rate. On the fourth pack, when100mls had been transfused, she developedacute dyspnoea, respirations were 40/min, andshe desaturated to 60% on room air andrequired O2 therapy. Initially TACO wasconsidered unlikely as the volume transfusedwas only about 700mls and the ConsultantHaematologist was asked to review thepossibility of TRALI. The X-ray showed rightsided atelectasis and a pleural effusion and thepre and post transfusion NT-pro BNP levels weremarkedly elevated (pre 1118, post 4897). In viewof the patient’s small size, rate of transfusion, X-ray findings and raised BNP, a diagnosis ofTACO was made. In addition SD plasma is apooled product which has not been convincinglyimplicated in TRALI.Reactions occurring in patients as a result of anerrorTACO was reported to have occurred following anerror in seven (18%) cases. Human error was citedas a cause of error in all seven cases and includedthe failure of the caregiver to adhere topolicies/procedures, to co-ordinate/communicateand to monitor the patient’s haemoglobin betweenunits.SAR Table 8 Breakdown of errors resulting inTACO (n=12)Human failureSystem failureFailure to adhereto policies/procedures 4 Culture 2Knowledge 1 Policies/procedures 1Co-ordination/communication 2Verification 2Total 9 3System failure was noted to be the cause of error inthree of these cases. In two cases the culture withinthe hospital contributed to the error and in the thirdcase lack of detail in the policy led to the error.SAR Case History 8 (TACO):A low weight (52kg) elderly female patient withPR bleeding admitted as an emergency wascommenced on IV fluids at 60ml/hr. Herhaemoglobin was 3.9 g/dl. The patient had nohistory of underlying cardiac, respiratory or renaldisease and the only relevant past history wasdiverticulitis. She was not on regular diuretics.She was transferred to ICU where three units ofred cells were transfused uneventfully. Thefollowing morning her Hb was 8.2g/dl. She wasthen transferred to the ward where she receiveda further unit (4th unit) of red cells with noadverse outcome. On day three the patientreceived 4L of fluids as part of bowelpreparation for gastro-intestinal investigationsand the IV fluids continued at a rate of 60mls perhour. No fluid balance was maintained at wardlevel. Her Hb was 9g/dl and a further two unitsof red cells were prescribed by the consultantwho prescribed diuretic cover if necessary. Theunits were given late in the evening. During thefirst unit (5th unit) her blood pressure (BP)started to rise with a systolic of 170mmHg and adiuretic was given post transfusion. Although thepatient remained hypertensive and had atachycardia the second unit (6th unit) wascommenced without query by the nursing staff.After approximately 70mls had been transfused,the patient’s condition had worsened, theconsultant was then contacted and thetransfusion stopped. A junior doctor reviewedthe patient and found pedal oedema and bibasalcrepitations on examination. Diuretics andan anti-hypertensive were administered and thepatient made a complete recovery within 12hours. On review by the ConsultantHaematologist, the 5th and 6th units weredeemed unnecessary and a contributing factorto the overload. Both human and system failureswere implicated in this reactionHuman failures• Verification: Failure to assess the patient’sclinical condition prior to prescribing furtherunits.• Failure to adhere to policies: Hb check notcarried out between units.55

• Monitoring: Failure to document an accuratefluid balance.System failureCulture: Nursing staff did not question consultantprescription although the patient was clearlysymptomatic.Discussion and RecommendationsTACO is probably the commonest complication ofblood transfusion with an estimated incidence of1:200. Particular attention should be paid toidentifying patients at risk. High risk patients includepatients of low body weight, elderly, infants andchildren and physiologically compromised patientsespecially those with chronic anaemia, cardiac, renaland respiratory conditions.The recipient’s weight should be taken into accountwhen deciding the number of units necessary. A unitof blood will raise the Hb by 0.5-2gms depending onthe size of the patient, and the size of the unit, whichcan vary in size as much as 30-40% (Davenport 2005,Gilcher 2002, Hogman 2006). The size of red cellunits issued by the IBTS is between 230mls to350mls with a mean of 260mls. The Hb leveltherefore should be checked between units to avoidover transfusion, as a single unit may be sufficient.Rapid transfusion also has been attributed as a factorfor circulatory overload. The NBUG (2004)recommends an infusion rate of 2-4mls/kg/hr for redcells in non bleeding patients but slower rates maybe required in small patients or in patients who haveunderlying cardiac disease or chronic anaemia.Popovsky (2001) suggests that patients at risk ofTACO who are not haemorrhaging should betransfused slowly at a rate of 1ml/kg per hour withcareful input and output monitoring before, duringand after the transfusion.The majority of transfusions - 24 (61%) - in this reportwere prescribed over three to four hours, howeverfive (20%) of these were transfused between one tothree hours and one patient, with a history of acuterenal failure, received a volume of greater than300mls in fifteen to thirty minutes.TACO in Massive Transfusion:Four cases of TACO in young healthy femalepatients have been reported to the NHO between2000 and 2007. The recognition of two further casesof TACO in 2008 in patients with massivehaemorrhage and no history of underlying disease isof concern and accounts for 5% of TACO cases.56Estimation of blood loss is often difficult, and inbleeding patients, the Hb on its own may not be agood measure of rapid blood loss, as it may beartificially low if there has been over hydration withcrystalloid or colloid although the red cell mass maybe normal (Valeri 2006, Soni 2008).Key PointsParticular attention should be paid to patientswith underlying conditions which may increasetheir susceptibility to TACO. These include;• Elderly patients• Infants and children• Patients of low body weight• Patients physiologically compromisedparticularly those with a history of cardiacrespiratory or renal insufficiency or chronicanaemia.Transfusion should be on a unit by unit basis, with amedical assessment of the patient prior tocommencing transfusion and a Hb check/assessmentbefore administering any further component. Thisassessment should include a careful estimation ofthe patient’s hydration and cardiac status prior to thetransfusion and a thorough review of the patient’sfluid balance during the transfusion.There is a possible need for diuretic therapy as thiscan reduce the risk of TACO and may be necessaryfor those on regular diuretic therapy.In very low weight or at risk patients, it may beadvisable to transfuse units with an interval of 24hours between each unit, in combination with pretransfusiondiuretics. Some patients take as long as24 hours to readjust blood volume particularly inthose patients whose venous pressure is raised pretransfusion(Mollison et al 1997).Doctors and nurses across all specialities shouldreceive education aimed at the recognition andavoidance of TACO. In addition junior doctorsshould receive specific training in the area oftransfusion medicine to ensure safe and appropriatedecision making regarding transfusion andprescription of blood components/products.It is important that clinicians recognise that evenhealthy patients can develop circulatory overload inthe massive transfusion setting and fluid balanceshould be carefully monitored to avoidoverhydration/overload with components.

Transfusion Associated Dyspnoea (n=2) 2008Definition: Transfusion Associated Dyspnoea(TAD) is characterized by respiratory distress within24 hours of transfusion that does not meet thecriteria of TRALI, TACO or allergic reaction.Respiratory distress should not be explained bythe patient’s underlying condition or any otherknown cause.FindingsThis is the first year the NHO has collected thistype of reaction and within this group we receivedtwo reactions that, following review, werecaptured as TAD. Both patients were elderly andboth reactions were associated with red cells.SAR Case History 9 (TAD)In the first case which was originally submittedas an unclassified transfusion reaction, apatient with underlying renal diseaseexperienced a drop in O2 saturation with anassociated tachycardia within 15 minutes ofcommencing the transfusion. There was noevidence to suggest that the patient was inoverload, nor did they show signs of anallergic type reaction. The symptoms resolvedwithin one hour following administration ofoxygen therapy.SAR Case History 10 (TAD)The second case was originally submitted asTACO but as there were no changes on thepatients chest X-ray from admission, it wassubsequently reclassified. The elderly malepatient, who was admitted with an underlyingmedical condition, was commenced on a unitof RCC. One hour 15 minutes post thetransfusion, the patient became acutely shortof breath, distressed and agitated. A diffusewheeze was noted. It was not considered anallergic type reaction. The patient was treatedwith nebulisers and IV medication andsubsequently required CPAP. The CPAP wasdiscontinued six hours later and the patientmade a complete recovery.Clinical OutcomeBoth patients made a complete recovery.57

Transfusion Related Acute Lung Injury 2008Transfusion Related Acute Lung Injury (TRALI) isone of the leading causes of transfusion relatedmortality.The NHO has adopted the Canadian Conferencedefinitions which divides TRALI into TRALI andPossible TRALI (Kleinman et al 2004).TRALI is characterised by the following• Acute onset of symptoms• Hypoxemia SpO2

tachycardia of 110/min, frothy sputum andblood stained secretions in her mouth. Heroxygen saturations disimproved (94% on100% O2). She was re-ventilated and givenfrusemide 40mgs with no noticeableincreased diuresis. A chest X-ray at this timeshowed bilateral perihilar alveolarconsolidation consistent with pulmonaryoedema, shock lung or aspiration. Her centralvenous pressure was 20 and remainedbetween 15-20 over the next eight hours. At08.00hrs on day 3 she was in a positivebalance of 2,396 millilitres. As her weight wasapprox 44 kg this was a considerable positivebalance. She received further doses ofdiuretic between day 3 and day 6. Her chestX-ray on day 4 showed some improvementcompared to the X-rays of day 2 but shecontinued to require ventilation until day 9.The absence of HLA antibodies in the donorsand the clinical features suggested TACO.Against TACO was the failure to respond todiuretics and the long period before recovery,and after discussion with the reportingphysicians, the case was collected as possibleTRALI.Differential diagnosis of Possible TRALIThe findings of dyspnoea, hypertension,tachycardia, positive fluid balance and chestX-ray changes were consistent with the ISBTdefinition of TACO. There was, however, noevidence of a diuresis followingadministration of a diuretic. BNP pre or posttransfusion levels which might be helpful andtroponin levels for cardiac ishcemia were notperformed. The electrocariodgram (ECG)showed no significant changes.Results of follow-up donor investigationsThere were three donors whose units weretransfused within the six- hour time frame ofthe current definition of TRALI or possibleTRALI. There was one male donor with noprevious history of transfusion and thereforenot investigated further. Both female donorswere recalled and there was no evidence ofeither HLA I or II or granulocyte antibodies ineither donor. Concurrently, HLAinvestigations of the patient were alsonegative.59

Suspected Transfusion TransmittedInfection2008The NHO collects and investigates reports of:• All STTI viral infections relating to bloodcomponents which have been transfused afterthe introduction of mandatory testing for thatvirus.• Viral infections not covered by mandatorytesting, e.g. hepatitis A virus, CMV andparvovirus, but which are suspected to beassociated with a blood transfusion.• The NHO also collects and investigates reportsof TTI – bacterial infections and parasiticinfections.Infections presenting weeks, months or years after atransfusion are termed post-transfusion infections.Bacterial or parasitic infections are usuallyassociated with acute symptoms and come toclinical attention soon after transfusion. Viraldiseases, however, may not be associated with anysymptoms until some years later. Therefore, reportsreceived within this category are not necessarily theresult of components transfused during thisreporting year.These reports of STTI may be due to the transfusionof an infected or contaminated unit, but equally,infection may have been acquired from anothersource. Investigation of markers of infection in animplicated donation, or in subsequent samples fromthe donors of implicated donations, can confirmtransfusion as the probable cause of infection, oridentify the need to investigate other possiblesources (SHOT, 1999). Such investigations mayinvolve microbiological testing of many donors andmay take many months to complete.• A post transfusion infection is confirmed astransfusion-transmitted once investigations arecomplete and the following criteria are fulfilled:(SHOT, 1999)infection prior to the transfusionand, either• A donor who had evidence of the sametransmissible infection donated at least onecomponent received by the infected recipientor• At least one component received by theinfected recipient was shown to have beencontaminated with the same infectious agent(SHOT 1999).FindingsIn 2008 eight reports of STTI were collected by theNHO. A further case of possible rubella transmissionwas reported but this case did not progress.SAR Table 9 STTI collected in 2008 (n=8)Type of STTInTransfusion transmitted bacterial infection 6Transfusion transmitted viral infection (HBV) 1Transfusion transmitted viral infection (HIV) 1STTI - Bacterial (n=6)Bacterial infection remains a rare, but seriouscomplication of transfusion, particularlyassociated with platelets which are stored at 20°C(Stainsby et al, 2006). The IBTS has introducedbacterial screening of all platelets before issuingand the diversion of the first aliquot of the blooddonation into the blood testing pouch which aremeasures which have been shown to reduce the riskof bacterial contamination (McDonald 2006).A recall of a component by the IBTS due to apositive bacterial culture (Bactalert) where thepatient has a reaction or is put on antibiotics iscollected by the NHO as a possible TTI.60• The recipient had evidence of infectionfollowing the transfusion, with no evidence of

FindingsSix cases of possible bacterial infection werereported. Four cases involved red cells and twocases were associated with pooled platelets. Fourwere considered possible bacterial infections andtwo were considered unlikely.Possible bacterial TTI were reported in two patientswho developed febrile reactions associated withtransfusion and who had positive blood cultures, butwhere the pack was not cultured (Case 1) or culturednegative (Case 2). However, further investigations athospital level suggested that these results were dueto contamination during the blood culturing processand bacterial infection was considered unlikely.These reactions were probably FNHTR.In another two cases, both involving RCC wherebacterial infection could not be excluded, acoagulase negative staphylococcus was identified inthe first patient, but the segment line from the packwas cultured with no growth (Case 3). In anothercase (Case 4) a pantoea species was cultured fromthe patient, (who had severe drug inducedneutropenia) the pack and the administration set.However, the segment line cultured negative.Two reports of possible bacterial infection involvedcases where neither patient had a reaction but werecommenced on antibiotics by the clinical teambecause of a report of an unconfirmed positivebacterial screening test from the IBTS (Cases 5 and6).SAR Table 10 STTI - Bacterial (n=6)CaseImplicatedNo. Age Gender Component Organism Outcome Imputability1 Adult Male RCC 1.Streptococcus Gordoni. FNHTR. Bacterial Unlikely(31-50 2.Coagulase negative Culture -consideredyears) staphylococcus identified most likely ain the patient.contaminantPack not cultured2 Elderly Male RCC Gram positive bacillus FNHTR. Bacterial Unlikely(70+) identified in the patient. Culture -consideredPack and segmentmost likely aline culture was negative contaminant.3 Elderly Male RCC Coagulase negative Febrile reaction Possible(70+) staphylococcus identifiedin the patient. Pack notcultured howeversegment line wascultured with no growth.4 Elderly Female RCC Pantoea species cultured Febrile reaction Possible(70+) from patient, pack and in patient withadministration set. Culture severe neutropenianegative from segment line.5 Adult Male Pooled Coagulase negative Unconfirmed positive Possible(31-50 Platelets staphylococcus Bactalert. Patientyears)had no reaction butwas commenced onantibiotic therapy6 Adult Female Pooled Coagulase negative Unconfirmed positive Possible(31-50 Platelets staphylococcus Bactalert. Patient hadyears)no reaction but wascommenced onantibiotic therapy61

Clinical OutcomeAll the patients made a complete recovery.STTI – Viral (n=3)Findings:Three cases of viral infections were reported to theNHO in 2008 of which two progressed. The firstcase, a case of HBV (Case 1) involved multiplecomponents and a large number of donors.Following investigation, TTI was consideredunlikely.The remaining two cases involved RCC only.In case 2, a case of HIV, the three donors returnedand retested HIV negative and TTI was excluded.In case 3, a male patient developed rubella eightdays following transfusion. Transfusion transmissionof rubella has never been reported and oninvestigation of the donors, both donors werefound to be immune to rubella prior to donating.Closer questioning of the recipient indicated hehad been exposed to rubella some days earlier.This case did not progress.SAR Table 11 STTI - Viral 2008 (n=3)Case Serious Age Gender Transfusion Components Donors Comments ImputabilityNo. Adverse Date ImplicatedReaction1 HBV Adult Female 1-Apr-08 RCC, SD 98 91 donors Unlikely(51 - 70 Plasma, returned andyears) Apheresis retested HBVPlatelets,negative.Pooled7 donorsPlateletstested HBVnegative onarchive samples.2 HIV1/2 Adult Female 9-Apr-08 RCC 3 All three donors Excluded(31-50 returned andyears)retested HIVnegative.3. Transfusion Adult Male 23-Oct-08 RCC 2 Recipient had Excludedtransmitted (31-50 exposure toviral years) rubella priorinfection -to transfusionOther-Both donorsRubellawere immuneto rubella priorto donating.Case Did NotProgress62

Serious Adverse Reactions occuringas a result of an error 2008This is the first year that this data has beenavailable for all the reactions reported. In 10 of the143 reactions reported the reaction occurred as aresult of an error (SAR Table 12).SAR Table 12 Reactions occurring as a result of an error (n=10)Type of Reaction Number Error DiscoveredAA 1 Failure to administer prescribed premedicationImmunological haemolysis 1 Incorrect component issued - Fy a positive unitdue to other allo-antibodyaccidentally substituted for a Fy a negative unit(Acute < 24 hrs)and transfusedUnclassified SAR 1 Incorrect volume of RCC transfused due to acommunication errorTACO 7 Transfusion time too short n=3Prescribed and transfused too quickly n=1Transfusion based on an old Hb result and noprophylactic diuretic administered n=1Unnecessary Transfusion n=1Units administered too close together in a patientwith underlying cardiac disease n=1The majority of the transfusions were administeredin the general ward setting (7) and one each in A&Edepartment, day ward and neonatal unit. All theerrors occurred in the clinical area with theexception of one case where the patient had anAHTR due to an error in the HBB.An analysis of the findings showed that in themajority of the cases several factors contributed tothe errors. The causes of the errors were identifiedas follows: human error was cited in all ten caseswith system failures also cited in five of these cases.The type of human error identified is outlinedbelow in table 13.SAR Table 13 Human Failures identified (n=17)Human FailuresKnowledge 2Co-ordination/Communication 4Verification 3Failure to adhere to policies/procedures 4Monitoring 1Slip 2Carrying out task incorrectly 1In eight of the cases, several human failuresresulted in the error with coordination/communicationand failure to adhere topolicies and procedures listed as the most commoncauses of human error.n63

In five of the cases, system failures were also listedas a cause of error with one case listing bothmanagement priorities and culture as contributingfactors in the error.SAR Table 14 System Failures identified (n=6)System FailuresnPolicies/Procedures 2Management Priorities 2Culture 2Clinical OutcomeThe outcome was documented in nine of the tenpatients. Six patients recovered fully, two patientshad minor sequelae and one patient diedunrelated to the transfusion.SAR Table 15 Clinical Outcome (n=9)Clinical OutcomenComplete Recovery 6Minor Sequelae 2Death 1 Unrelated totransfusion64

Paediatric Serious Adverse Reactions2008“The acute side effects of transfusion may be greaterfor small children than for adults, as a single unit oftransfused blood with the potential to cause harm,may represent a much greater proportion of theirblood volume than that in an adult” (New 2006 P.2)This year 19 (13%) out of a total of 143 reactionsoccurred in paediatric patients. Four reactionsoccurred in neonates (

were to cryoprecipitate, both unclassifiedreactions were associated with RCC and the oneFNHTR was associated with the transfusion ofgranulocytes.Clinical OutcomeA clinical outcome was given in 17 of the cases.Fifteen patients made a full recovery, onepatient had minor sequelae and one patient hadserious sequelae but subsequently recovered(see SAR Case History 4).Paediatric SAR Table 3 Clinical Outcome forpaediatric reactions (n=17)Clinical OutcomeComplete Recovery 15Minor sequelae 1Serious sequelae 1n66

Incorrect Blood Component Transfused(IBCT)Serious Adverse Events (SAE) 2009FindingsIn 2009, 157 IBCT/SAE related to bloodcomponents and blood products were acceptedby the NHO, representing 58% of analysedreports.Thirty five reports related to blood products(factor concentrates and anti-D) and these areseparately assessed on pages 89 and 91.IBCT/SAE associated with blood componentsand SD plasmaThe NHO analysed 122 reports relating to bloodcomponents and SD plasma. These reports weresubmitted from 41 reporting establishments• Elderly patients aged over 70 years wereinvolved in 20% of reports. Twenty one (10%)reports involved paediatric patients

Mandatory SAE reports under Directive2005/61/EC (n=46) 2009Mandatory SAEs do not include errors related toclinical aspects of blood transfusion practice, but arerelated to activities in blood establishments (BE) andHBB 6 . These can be deviations in testing, storage,materials, distribution or other aspects of BE/HBBactivity.The format of mandatory event collection is basedon the EU Directive therefore some cases arecaptured in different categories to those of theNHO. This means that some SAE /IBCT areIBCT/SAE Table 11: Mandatory SAE 2009 (n=46)described under different headings in the narrativeSAE and IBCT and Paediatric sections of the report.FindingsForty-six mandatory SAEs were reported to theCompetent Authority in 2009 (IBCT/SAE Table 11).This table is in the format set out in EU Directive2005/61/EC. Mandatory SAEs were 38% of allSAE/IBCT a 13% decrease compared to 2008.Thirteen (52%) SAEs involved paediatric patients,and are described in the Paediatric section on page84. None involved sequalae for the patient.Deviation Total Component Specificationn (%) RCC Platelet FFP Cryo Human Error OtherTesting of donations 3 (7) 3 3Storage 7 (15) 7 6 1Materials 1 (2) 1 1Transfusion of an 13 (29) 11 2 13incorrectly labelledcomponentNon-irradiated/andnon- CMV negativecomponents transfused * 6 (13) 6 1 5 1Incorrect ABO Grouptransfused (no reaction) 1 (2) 1 1Transfusion of otherantigen incompatibleRCC (no reaction) 3 (7) 3 3Incorrect componenttransfused (no reaction) 8 (17) 4 3 1 8Incorrect Rh grouptransfused 1 (2) 1 1Transfusion of expiredcomponent 1 (2) 1 1Other 2 (4) 1 1 1 1Total 46 (100) 39 6 1 1 42 4686Mandatory SAEs related to the quality and safety of blood components are reportable to the NHO under legislative regulations detailed in the NHO handbook.(*one case involved the transfusion of multiple components).

Testing of Donations (n=3)In one case, the pre-transfusion crossmatchidentified anti-c and anti-E antibodies and anunidentified antibody. Five units of c negative Enegative red cells were ordered from the supplycentre. However, during the transfusion of the firstunit the patient became pyrexial and developedtachycardia and dyspnoea. A check of on-call workshowed that the unit involved had been positive oncrossmatch but that the crossmatch card had beenincorrectly read as negative and the unit issued. Posttransfusion investigations showed no evidence ofhaemolysis or additional red cell antibodies,although the patient also had HLA antibodies. Thereaction was considered unlikely to be related totransfusion.The second case involved failure of the controlsduring crossmatch of the pre-transfusion sample.This went unnoticed and the unit was issued andtransfused. Root cause analysis attributed this tohuman error. Post transfusion investigations foundthe unit to be compatible. These cases werecategorised as “Other” when reported to the NHO.The final case is described in IBCT/SAE Case History20.IBCT/SAE Case History 20This case involved a post-menopausal female whowas typed as ORhD positive and issued with aunit of ORhD positive red cells. Followingcommencement of the transfusion, the patientinformed the nurse she was RhD negative and thetransfusion was stopped. Investigations found norecord of this patient on the LIS, but a furthersearch revealed a historical record with a differentspelling of the surname. This case wascomplicated by an anomalous result of a weak Dduring RhD typing of the pre-transfusion sample.The pre-transfusion and repeat samples weresubsequently referred to the reference centrewhich confirmed the patient’s blood group as RhDnegative. The patient suffered no sequelae.Incorrectly Stored Components (n=7)Seven (15%) cases were implicated in storage errors.All involved red cells, six were due to human errorand one to a systems failure. One case is describedin the Paediatric section on page 85 underIncreased donor exposure due to problems withpaedipacks.Materials (n=1)One case was reported as a deviation in ‘materials’.This led to the recall of a paedipack by the supplycentre due to a problem with materials used in thetesting. The baby was not commenced on anyantibiotic treatment, but was exposed to anotherdonor. This is one of the cases detailed in thePaediatric section on page 85 in the categoryTransfusion of the Incorrect Component/Productleading to Increased donor exposure due toproblems with paedipacks.Incorrectly Labelled components (n=13)The largest category of mandatory SAEs (n=29) wereincorrectly labelled components, all of which wereattributable to human error. Eleven involved red cellsand 2 involved platelets. One case is described inthe paediatric section on page 86.• Seven cases involved transposition of labelswithin the same crossmatch of red cells, and inone case transposition of labels occurred duringlabelling of platelets. In two instances severalcomponents were issued simultaneously andsubsequently returned to stock. When anattempt was made to reissue the returnedcomponent, the LIS indicated it was alreadytransfused.• In another case, a unit of pooled platelets wasissued and transfused without a compatibilitylabel.• In a further case, an incorrectly labelled unit ofred cells was issued and transfused. When theMedical Scientist went to print a label, he usedthe last three digits from the pack to identify andselect the unit on the LIS. The label with anincorrect unit number was printed and attachedto the pack. The hospital label was not crosscheckedwith the supply centre label duringissue, collection or pre-transfusion check of thecomponent.• The final case involved a patient with autoantibodies for whom three units of red cells wereassigned. Two units were requested from thelaboratory and the two least incompatible unitsselected by the medical scientist. However thelabel was incorrectly applied to the third unit inerror.Non-irradiated/CMV Negative ComponentsTransfused (n=6)Six (13%) cases were captured in this category, all ofwhich involved red cells and one case also involved69

platelets. All were due to human error, and one casewas attributable to both human and system errors.Incorrect ABO Group Transfused (n=1)This case is described in the Paediatric section onpage 85.Transfusion of Other Antigen Incompatible RCC(n=3)Three (7%) cases were captured in this category, thefirst of which is described in the paediatric section. Inthe second case, a patient with a history of anti-Eantibodies required an urgent transfusion of redcells. The on-call scientist was unaware that Enegative units should have been selected based onthe historical record. Screening of the pre-transfusionsample was negative and the units were crossmatchcompatible but antigen negative units were notselected for issue. On retesting, a weak anti-E wasdetectable in the sample. It is unknown if the redcells issued were E antigen positive. The patientsuffered no sequelae.IBCT/ SAE Case History 21In the final case a patient required an urgenttransfusion of red cells for acute gastrointestinalhaemorrhage. An on-call medical scientist whodid not normally work in transfusion had difficultyin resolving the crossmatch in this case. On initialtesting the antibody screen was negative and 2units of red cells were issued. The scientisthowever, then noticed an anomaly in the resultsand repeated the crossmatch and antibody screenwhich was positive. A more senior scientist wasthen called who retested the sample and set upan antibody identification panel. This indicatedthat most probably anti-Fy a was present and Fy apositive components had been issued. It wasdiscovered that the first unit had already beentransfused. The patient suffered no sequelae.Incorrect Component Transfused (n=8)In eight (17%) cases the incorrect component wasissued. All were due to human error. Six casesinvolving paediatric patients are described in thePaediatric section on page 85.• One case involved the issue of apheresisplatelets instead of pooled platelets, to a patientwith a history of previous SAR.• In a further case, a patient was grouped as ARhD negative, but the HBB issued an O Rh Dnegative red cell unit marked as high titre anti A70and anti B which was not noticed by the medicalscientist .Incorrect Rhesus Group Transfused (n=1)In this case, two units of red cells were requested foran Rh D negative male patient. A busy on-callmedical scientist did not heed a warning on the LISsystem that a Rh D positive component was beingissued to a Rh D negative patient. The error wentunnoticed during issue, collection and at the pretransfusionbedside check, and was only discoveredprior to transfusion of the second unit.Transfusion of an Expired Component (n=1)In this case, a patient was transfused with a unit ofred cells which were issued two hours prior to expiry.Clinical staff were not informed of this and thetransfusion was commenced two hours past theexpiry time. This was not detected during the pretransfusioncheck.Other SAEs (n=2)The remaining two incidents were captured underthe deviation ‘Other’. One case involved a patientwho received an insufficient dose of cryoprecipitate.Four units of pooled cryoprecipitate wereprescribed, but due to a shortage of pooledcryoprecipitate single donor units were issued by thesupply centre. The hospital medical scientist wasunaware of this and issued four single donor units.As each unit of pooled cryoprecipitate contains fiveunits the patient received an inadequate dose. Thesecond case is described in the paediatric section onpage 85 under ‘Increased donor exposure due toproblems with paedipacks’.Cause of SAEsEU Directive 2005/61/EC states the cause of the SAEmust be investigated through root cause analysis(RCA). In addition, the NHO assesses each case forits potential to cause harm. Sixteen (35%) SAEs weredetermined to have a major potential to cause harm.These included storage errors (3), materials (1),testing (1) and labelling errors (1), issue of theincorrect ABO group of red cells (1), issue of theincorrect component (5), issue of antigenincompatible components (3) and other (1). Theremaining 30 (65%) mandatory SAEs were assessedas having moderate potential to cause harm.IBCT/SAE Table 12 shows a breakdown of errorcauses and their potential for harm. In someinstances there was more than one or more errorcause.

IBCT/SAE Table 12 Root Cause Analysis (RCA) and Risk of Potential for Harm for Mandatory SAEs(n=86*)Human Failure Major Moderate Systems Failure Major Moderaten (%) n (%) n (%) n (%)Failure to adhere to policies 8 (9) 16 (19) Policies/ 1 (1) 2 (2)& procedures.ProceduresVerification 2 (2) 4 (5) Design 2 (2) 3 (4)Knowledge 5 (6) 5 (6) Materials 2 (2)Carrying out task incorrectly 3 (4) 13 (15) Other 2 (2) 1 (1)Slip 6 (7) 4 (5)Co-ordination/ Communication 1 (1) 5 (6)Other 1 (1)Total 26 (30) 47 (55) 7 (8) 6 (7)(*RCA revealed more than one error cause for some cases)Follow-up ActionPreventative and/or corrective action was taken in26(65%) of the 46 mandatory SAE. Measures takenincluded the re-education and/or training of staffinvolved in transfusion practice, and the introductionof new policies stipulating procedures, updatinghospital LIS, for example to flag patients with specialtransfusion requirements.Key Point• Several mandatory SAEs were associatedwith on-call medical scientists with abusy workload.• Over half (n=13,52%) of all paediatricevents reported were mandatory SAEs.• The UK Transfusion Collaborative (2009)recommends that laboratories havesystems in place to ensure adequate skillmixes and staffing levels to ensure a safeand effective service to patients duringboth routine and ‘out of hours’ servicesRecommendations• There should be robust systems to detect andreconcile patients’ previous histories andtransfusion records (SHOT, 2009).• Antigen negative blood should be providedwherever the patient's condition allows, but in a‘Code Red’ type emergency bleed, it may benecessary to transfuse patients who haveantibodies with units from the emergency O RhD negative stock although they may not beantigen compatible as the risks due to delay intransfusion caused by trying to find compatibleblood out weigh the risks of a haemolyticreaction.• Hospital blood bank staff must be particularlyvigilant where patients have special requirementsand adhere to hospital policy to ensure thecorrect components are issued.• Clinical staff should be alerted if issuedcomponents are about to expire.71

Adverse Events in the Clinical Areas(IBCT/Non Mandatory SAE) (n=76) 2009This section, describes the findings on the mostfrequently reported adverse events occurring in theclinical area.Unnecessary transfusions (n=25)This captures adverse events where a patient istransfused with a blood component which was notrequired (NHO, 2007).There were 25 reports of unnecessary transfusions in2009, making up 20% of all SAE /IBCT reports,making unnecessary transfusion the highest singleclinical adverse event to be reported. This comparesto 15 reports in 2008 showing an increase of 10% onthe 2008 report.• Unnecessary transfusions involved red cells in 14cases , platelets in six cases (one report involvedboth SD plasma and platelets, and the platelettransfusion was deemed unnecessary) and SDplasma in five cases.• Twelve cases involved transfusion episodeswhere the entire transfusion was consideredunnecessary. In two cases, the transfusioninvolved multiple units and in these cases onlysome units were considered unnecessary.• Doctors were implicated in 88% (22) of caseswith prescription /request reported as the site oferror in 72% (18) of cases.• Knowledge deficits were identified in 56% (14) ofunnecessary transfusions predominantly involvingdoctors, but also involving other staff.Key Point• The high incidence of knowledgedeficits in unnecessary transfusionshighlights the important role ofeducation of clinical staff in theappropriate use of blood.Unnecessary transfusions were classified as follows.IBCT/SAE Table 13: Classification of unnecessarytransfusions (n=25)ClassificationTransfusion based on clinical decision notin conformity with guidelines 15Transfusion based on incorrect or absenthaematology result 9Other 1Transfusion Based on Clinical Decision not inConformity with Guidelines (n=15)Iron Deficiency Anaemia• Five cases of unnecessary red cell transfusion inthis category involved patients with irondeficiency anaemia who received transfusions.Four of these cases involved female patientsaged between 29-49yrs.• Two patients had iron deficiency anaemia dueto menorrhagia.- One patient had been referred by her GPfor investigation of a one year history ofmenorrhagia. Her pre-transfusion Hb was7.6g/l and her vital signs were stable.- The second patient was admitted withiron deficiency anaemia. She had a historyof menorrhagia and was also a vegetarian,and was not compliant with oralmedication. Her pre-transfusion Hb was6g/dl. Intravenous iron therapy was notconsidered.• Two other patients had other conditions withco-existing iron deficiency anaemia where atleast one of the units transfused wasconsidered unnecessary.- One patient was admitted for cardiacinvestigations. Her pre- transfusion Hbwas 7.2g/dl and she was transfused threeunits of red cells. Her post transfusion Hbwas 12g/dl. This patient was not on irontherapy.- The second case involved a 37 yr oldn72

patient with iron deficiency anaemiafollowing previous extensive bowelsurgery who received three units of redcells. A review at the hospital transfusioncommittee considered the final unitunnecessary. Intravenous iron therapywas not considered.• A fifth patient with iron deficiency anaemiawas given a second unit of red cells althoughthe pre-transfusion Hb was 11 g/dl.Key Points and Recommendations• Underlying anaemia has beenrecognised as a cause of unnecessarytransfusion and increased morbidity inpatients undergoing elective surgery. Arecent publication by the Network ofAdvancement of TransfusionAlternatives made recommendations ondetection, evaluation and managementof pre-operative anaemia (Goognoughet al 2010).• In 2009, five patients 7 who weretransfused unnecessarily had irondeficiency anaemia. Year on year, theNHO receives a number of reports ofunnecessary red cell transfusion inpatients with iron deficiency anaemia,and this is likely to represent significantunder reporting. An audit of patientswith iron deficiency anaemia managedin a tertiary care hospital identified thatapproximately 10% (3) of the studygroup were unnecessarily transfused,and that 30% (9) of patients had morethan 1 unit where iron therapy couldhave been given instead of additionalred cell transfusion (Egan et al, 2010).• Nutritional anaemia due to iron, folate,vitamin B12 deficiency respond rapidlyto appropriate haematinic therapy.• Intravenous iron preparations should beconsidered in cases where patientshave either poor tolerance of oralpreparations or there are complianceissues. It normalizes haemoglobin fasterand more reliably than oral iron(Ahmad, and Gibson,2006).• Ferric carboxymaltose although moreexpensive may deliver several potentialadvantages over other parenteral ironpreparations. It is feasible to administermuch higher single doses of iron overshorter periods of time, resulting in theneed for fewer injections to replete ironstores and correct iron deficiencyanaemia. It provides a rapid responsein haemoglobin with fewgastrointestinal side effects (Kulligg etal, 2008).Other unnecessary red cell transfusions• Two cases of red cell transfusion were not basedon best practice guidelines.• In one case a post–operative patient with aHb of 9.6g/l received two units of red cell for“nausea associated with a low Hb”. The posttransfusion Hb was 11.6g/l.• The second case involved a four unittransfusion to a 56 yr old patient whose pretransfusionHb was 7g/dl. His post operativeHb was 12g/l. The Hb was not checkedbetween units.Unnecessary SD plasma and platelet transfusionsassociated with procedures (n=7)All cases involved a lack of clinical knowledge ofeither the guidelines or the practical requirementsfor components pre-procedures.Lack of practical requirements for pre -procedure administrationIn four cases, patients were transfused bloodcomponents at the wrong time prior to plannedprocedures. In two cases, the procedure had notactually been arranged.• In one case, the patient received a unit of SDplasma for a liver biopsy which had not beenorganised.• In the second case, the haematologist asked toreview a medical patient with evidence ofbleeding and coagulopathy, ordered a platelettransfusion to be given prior to a plannedprocedure. Two units of platelets weretransfused but the procedure had not yet beenscheduled. Following review it was agreed thatwhile the patient required one unit of platelets7Final case reported as Transfusion based on Incorrect /Absent Haematology Result.73

for a concurrent clinical condition, the secondunit was unnecessary.• In two further cases, the procedure had beenorganised but was repeatedly postponed in onecase (IBCT/SAE Case History 22) and in thesecond, the patient was transfused on the dayprior to the planned procedure.IBCT/SAE Case History 22A patient with a platelet count of 12 x 10 9 / L wasscheduled to have a multilumen central lineinserted under radiological guidance. Hospitalguidelines state patients should have a plateletcount of 50 x 10 9 / L prior to procedures but amember of the X ray department ordered that theplatelet count should be 70 x 10 9 /L preprocedure. Three units of platelets were orderedand transfused. The post transfusion plateletcount was 88 x 10 9 / L. Following the transfusion,it was discovered that a specific introducer, wasrequired which was not available in the X Raydepartment and the procedure was postponed.On the next day, the patient’s platelet count haddropped to 58 x 10 9 , and the patient wastransfused a further unit of platelets again onadvice of the X ray department. On this occasionalthough the introducer was available, thepatient’s radiology slot was taken by anemergency patient and the procedure was againpostponed. On the following day, the patient’splatelet count was 49 x 10 9 / L and the patientwas again transfused with a further unit ofplatelets. The HVO became aware of the problemthrough the weekly platelet audit and afterdrawing the attention of the X ray staff to theguidelines, the procedure was then carried out.This case illustrates how failure to adhere toguidelines can delay procedures for patients andsignificantly increase donor exposure and costs.• The second case involved a patient with a lowplatelet count who was seen by a haematologistprior to a liver biopsy. The haematologistordered a unit of platelets to be given prior tothe procedure but the clinical staff did not realisethat the platelets should only be transfusedshortly before the procedure as the platelets onlylast for a short time in the circulation and thepatient was transfused on the day before theprocedure. The patient then required a secondunit of platelets on the following dayimmediately prior to the procedure.Lack of knowledge of pre-procedure guidelines• A patient with a normal coagulation screenreceived an unnecessary unit of SD plasma priorto a guided ultrasound on instructions of aconsultant not routinely used to prescribingblood components.• In another case, a patient under care of bothsurgical and cardiology teams with a plateletcount at 384 x 10 9 /L received an unnecessaryplatelet transfusion prior to plannedneurosurgery. This patient was not takingmedication which would potentially inhibitplatelet function such as aspirin or clodiprogel.This patient also received two units of SD plasmafor a slightly raised INR (1.7) for reversal ofwarfarin, where vitamin K might have beensufficient.• In one case, two units of platelets wereprescribed and transfused to a patient with aplatelet count of 97 x 10 9 /L prior to a liverbiopsy, to raise the platelet count to 100 x 10 9 /L.The platelets were ordered by a junior doctorwho was unaware of hospital guidelines onclinical use of platelet transfusions and did notseek advice from senior clinical colleagues.Key Point• Hospital staff who work in specialisedareas may not be aware of hospitalguidelines for components and mayneed to have specially targetededucation sessions.• Platelets should be given within one totwo hours (or as close as is practical toallow for measurement of posttransfusion values) prior to the plannedprocedure.• The nature of hospital activity canimpact on planned procedures. Medical,nursing and other clinical teams caringfor patients along with HBB staff shouldclosely monitor the clinical activities andco-ordinate care to ensure patients donot unnecessarily receive bloodcomponents, where procedures arecancelled or postponed.74

Other unnecessary transfusions associated withinadequate knowledge (n=2)Two unnecessary SD plasma transfusions wereadministered for bleeding.• A patient who had a post partum haemorrhagewho was stable in the recovery room, receivedtwo units of SD plasma. At this time thecoagulation screen was within normal limits, butthe prescribing anaesthetist recollected posthaemorrhage “oozing”.• IBCT/SAE Case History 23In this case, a young male patient undergoingan elective revision of a hip replacement had asample sent to the HBB prior to surgery.Antibodies were identified on the pretransfusionsample and the HBB informed thesurgical team that compatible blood would notbe available and asked the team to delay theprocedure. A clinical decision was made toproceed with the surgery. The patient bledand was transfused six units of SD plasmaalthough emergency O Rh Negative bloodwas available. The rationale for this decisionwas not clear. The patient’s pre-operativecoagulation screen was normal.Key Point• The role of the HBB, not only as asupport service, but as an activeparticipant in delivery of safe patientcare, is made clear in this case but thiswas not recognised by the surgical teamcaring for the patient.Transfusion Based on Incorrect or AbsentHaematology Result (n=9)Five cases involved unnecessary red cell transfusionsfollowing a failure to verify results.• A transfusion was administered on the basis ofan incorrect result from a blood gas analyser.This was not an emergency transfusion.• A failure to verify that the Hb result was the mostrecent result led to a patient receiving anunnecessary red cell transfusion. This wascomplicated as haematology testing was carriedout at another site which resulted in a delay inthe availability of current results.• Incorrect transcription of results by an admittingdoctor resulted in an unnecessary red celltransfusion.• A patient with a Hb of 10 g/dl received anunnecessary red cell transfusion. The juniordoctor who prescribed the unit of red cells,failed to verify the patient’s correct Hb, when hewas told by a nurse it was 7.8g/dl.• An elderly female patient, with cellulitis and irondeficiency anaemia, was prescribed two units ofred cells on separate days. Prior to the secondtransfusion, her pre-transfusion Hb was 11 g/dland this result was not checked and the secondunit was transfused 8 .Three unnecessary transfusions occurred followingsampling errors.• Two unnecessary red cell transfusions occurredfollowing sampling errors made by medical staffdrawing blood for Hb testing. In one case, asample was taken from a peripheral vein whereintravenous fluids were infusing. In the secondcase, a doctor flushed a central line prior totaking the sample for FBC, discarded the FBCand inadvertently sent the diluted sample to thehaematology laboratory.• In the third case, SD plasma was transfused onthe basis of an incorrect coagulation result froma haemodiluted sample. In this case, aninexperienced phlebotomist took the sample.While the error was identified by the coagulationlaboratory and a further sample for coagulationscreen had been requested by the laboratory,the result had not yet been phoned to theclinical area, prior to the administration of SDplasma.In the final case, an oncology patient about tocommence chemotherapy was prescribed plateletson the basis of a platelet count of 13x10 9 /L. Whilethis result was available on the LIS, it had not beenvalidated. The validated result was 57 x10 9 , but bythis time, the platelets were transfused.Other (n=1)One case involving a two month old infant involvedmultiple clinical failures to verify results and decision8Previously referred to under unnecessary transfusion in iron deficiency anaemia75

making which deviates from best practice guidelines.This case is described in the paediatric unnecessarytransfusion section on page 86.Key Points• Three cases of unnecessary transfusioninvolved either a delay in posting currenthaematology results to the LIS or whereposted results were not validated.Laboratories should ensure thatvalidated results are available to clinicalareas in a timely manner.• A further three involved transfusionbased on results from haemodilutedsamples, a recurring problem.• Near patient testing may be necessary inemergency settings. Where this is used,maintenance and validation ofequipment as well as ongoing trainingand competency of clinical staff must beensured (NHO, 2006). Hb results whichsuggest a need to transfuse should bere-checked in the laboratory prior totransfusion wherever possible.Incorrect component/product transfused (n=21)This category captures incidents where the patientrequired a blood component/ blood product but themost appropriate one was not administered.• In 2009, the NHO received 21 reports in thiscategory, eight relating to SD plasma, one toFFP, eight to red cells and four to platelets.• Prescription/request and the HBB were reportedas site of error for 18 SAE/IBCT, equallyimplicating doctors and medical scientists. Eightof these cases were mandatory SAE and arereported separately on pages 70.Inappropriate use of SD plasma for reversal ofwarfarin (n=7)Seven reports related to the use of SD plasma forreversal of warfarin, where it would have been moreappropriate to use PCC.• Transfusion of one or two units was reported infive of these cases. Unless the patients wereextremely small, this was unlikely to be atherapeutic dose.Interestingly, two cases of the seven cases reportedas unnecessary transfusion involved patientsreceiving sub-therapeutic doses of SD plasma.• In one case, SD plasma was administered outsideroutine working hours when the junior doctor didnot wish to disturb the Consultant Haematologistto prescribe the PCC.Transfusion of the incorrect component or SDplasma, which was inappropriate to needs ofpatients (n=9)In three cases the incorrect red cell component wastransfused.• A nurse collected two emergency group O Rh Dnegative units from emergency stock, when fullycrossmatched units were available. While theHBB telephoned the ED that cross matched unitswere available, this information was not passedon to the nurse looking after this particularpatient..• Two of these cases were mandatory SAE and arereported separately in paediatric and mandatorysections on pages 70 and 85.There were four cases where the incorrect plateletwas transfused to patients.• Two cases involved transfusion of pooledplatelets to paediatric patients instead ofapheresis platelets.• A doctor ordered HLA matched platelets fromthe supply centre, but did not inform the HBB.When the clinical area ordered platelets from theHBB, new HLA matched apheresis platelets wereissued and transfused.• An adult patient with history of previous SARshould have received pooled platelets, but dueto an error occurring in the HBB, was transfusedapheresis platelets and this was reported as amandatory SAE on page 70.In one case, the HBB issued FFP instead of SDPlasma to a paediatric patient. This was reported asa mandatory SAE IBCT/SAE Table 11 and is listed inthe Paediatric Section.In the final case, universal SD plasma (which remainsunlicensed in Ireland) was issued instead of groupspecific SD plasma. This was reported as a paediatricIBCT/ SAE on page 85.76

Transfusion of red cells to paediatric patientsleading to unnecessary donor exposure (n=6)There were six cases where paediatric patients wereexposed to red cell transfusions from additionaldonors due mainly to errors occurring in the HBB.These are reported in the paediatric section on page85.Failure to transfuse special requirements (n=18)This category captures incidents where a patientwho required special requirements such as CMVnegative and/or irradiated blood components, didnot receive the required components.In 2009, the number of reports in this categoryincreased from seven reports in 2008 to 18 reports in2009, accounting for 8% of all IBCT/SAE reports in2009.• The majority of patients were adult patients.Eight reports (40%) in this category related topatients within the category 51-70 yrs and tworeports involved infants (1-12 months).• Fourteen of the SAE/IBCT occurred atprescription/request where doctors did notprescribe/request special requirements forpatients. Four errors occurred in the HBB, whenCMV negative and/or irradiated componentswere not issued to the patient. Six reports wereclassified as mandatory SAE.IBCT/SAE Table 14 contains a breakdown of thereports of failure to transfuse patients with bloodcomponents which were not CMV negative orirradiated.IBCT/SAE Table 14: Breakdown of the reports byindication for transfusion of special requirementsn=18Reports received in 20091. Transfusion of blood components whichwere not CMV negative and / irradiatedto patients pre and post solid organtransplants 9 . 52. Transfusion of blood components whichwere not irradiated to patients withHodgkins and non Hodgkin’s lymphoma. 13. Transfusion of blood components whichwere not CMV negative and irradiated topatients with Non Hodgkin’s lymphoma. 24. Transfusion of blood components whichwere not irradiated to patients with ahistory of either Hodgkin’s or NonHodgkin’s lymphoma. 45. Transfusion of blood components whichwere not CMV negative and / irradiated toa patient with congenital deficiency. 16. Transfusion of blood components whichwere not CMV negative to pregnantpatients. 27. Transfusion of blood components whichwere not CMV negative to a patientwith HIV. 1N8. Transfusion of blood components whichwere not CMV negative and / irradiatedto a patient with acute lympocyticleukaemia. 19One patient with nephritic syndrome was prescribed but did not receive irradiated components. The reason for this prescription is assumed tobe the potential for future renal transplant.77

• Some reports received in this category related totransfusions which did not conform with localguidelines. Hospital and laboratory policies arevery often broader that the published guidelinesfor transfusing CMV negative or irradiatedcomponents, such as those from McClelland(2007- See Appendix 1), AABB (2008 – for clinicalindications for irradiated components)) and morerecently the BCSH (2010).• The objective of these blanket policies is toensure that patients at risk receive bloodmeeting their specific requirements. However, itappears clear that national guidelines onappropriate use of irradiated and CMV negativeblood components are required. In this context,hospitals are likely to revise their local policies.Key Points• Reports relating failure to transfusecomponents with special requirementsshould only be submitted to the NHOwhere there is clear evidence for theiruse. Hospitals are encouraged to reviewagainst expert published guidelines.• There is a need for a national policy onappropriate use of irradiated and CMVnegative blood components.• Studies have demonstrated that staffinvolved in adverse events may becomevery distressed and this impact may belong lasting. Very often there may belimited resources available to supportstaff (Wu, 2000; Scott et al, 2009).Therefore it is important that reports ofserious adverse events relate to actualevents and not to non conformanceswith local hospital policies. In caseswhere a hospital policy specifies specialrequirements outside expert guidelinessuch as those from BCSH, it isrecommended these reports should bemanaged as non conformances in thehospital quality system, and notreported as SAE/IBCT.Other (n=21)There were 21 non mandatory SAE/IBCT reported inthe category of “Other”. These all occurred in theclinical area.• Nineteen occurred at administration stage of thetransfusion process• Two occurred at prescription /request.IBCT/SAE Table 15: Classification of nonmandatory SAE/IBCT reported as “Other” (n=21)Report Description of events nIncorrect • Red cells transfused greater 13transfusion than 6 hourstime• Red cells transfusedvery quickly in patients atrisk of developing a SAR.Incorrect • Unit transfused using a 6transfusion standard fluidadministration administration set.setSAE affecting • Packs punctured 2integrity of at administration.packDiscovery of SAE/IBCT (n=122)This section of the report will focus on discovery ofboth mandatory and non mandatory events.IBCT/SAE Table 16: Who discovered SAE/IBCT?(n=122)Who discovered error?Haemovigilance Officer 43Medical Scientist 50Doctor 7Nurse 20No information available 1Unclear 1NSAE/IBCT were discovered by clinical and laboratorystaff working in all aspects of transfusion.• As in 2008, the majority of SAE/IBCT werediscovered by medical scientists in the HBBduring recheck of on–call work or at next testingevent. Medical Scientists working as HVOs, in the78

supply centre and in the haematology laboratoryalso discovered errors.• Thirty five (80%) SAE/IBCT discovered by HVOwere clinical events.• Almost 10% (20) SAE/IBCT were discovered byclinical nursing staff following the transfusion. Allof these events were related to a prolongedtransfusion time.• Seven SAE/IBCT were discovered by doctors.These were unnecessary transfusions, transfusionwhere the most appropriate component was nottransfused and/or where special requirementswere not transfused.Stage of the transfusion process where theadverse event occurred (n=122)This section of the report will focus on the stages ofthe transfusion process where mandatory and nonmandatory events occurred.IBCT/SAE Table 17: Adverse events by stage inthe transfusion process flow (n=122)Work processNSampling 3Prescription Request 44Laboratory Processing - Other 3Laboratory Processing - Blood Transfusion 35Storage 3Collection 9Distribution 1Administration 20Supply Centre 1Other 3Totals 122As in previous years including 2008 , prescription/request is the stage of the work process where mostadverse events (n=44) occur, followed by thoseoccurring in the HBB (n=35) and at administration(n=20).IBCT/SAE Figure 8: Where did the SAE /IBCTwith a high risk assessment occur? (n-122)45 -40 -35 -30 -25 -20 -15 -10 -5 -032Sampling-4421PrescriptionRequest-3LaboratoryProcessing-Other-LaboratoryProcessingBloodTransfusion-Also as in 2008, most high risk events occurred atprescription/request. However, the events whichoccurred at distribution and at the supply centrealong with all events which were classified as Otherwere high risk to patients.Overview of causal analysis –Root cause ofeventsWhy did the SAE/IBCT occur?13511This section of the report provides an overview ofthe causal analysis of both mandatory and nonmandatory events occurred.3Both human and system errors contributed toSAE/IBCT reported in 2009. The cause of error wasreported in 121 cases. The reason for the error didnot become clear in one case of an unnecessarytransfusion due to a delayed follow up as the eventwas only discovered during investigation of asuspected transfusion SAR.System ErrorsSystem errors were reported in 22 cases, and morethan one error was reported in two cases (n=24system errors).----Storage Collection Distribution Administration SupplyCentreTotal SAE194High Risk SAE1120411-3Other3-79

IBCT/SAE Table 18: System errors leading to events (n=24)Classification N Commentof system errorLack of policies and procedures 7 • Incomplete or no clinical policies on blood administration viablood warmer or in a specialist unit• Incomplete or no laboratory policies on specific steps e.g.pre/post cross match checks transfusion checks, aspects oftemperature monitoring.• No details on one caseDesign 2 • No alert for special requirements on LIS• Display of pending results or delay in displaying current resultson haematology LISMaterials 3 • Pack leaked in clinical area• Extremely similar packaging of both blood and IV fluidadministration set.• Testing difficulties with platelet additive solution• Pipette failureManagement 2 • Clinical management of patient who received possibly twounnecessary transfusions• No system in hospital for haemovigilance and transfusioneducational updates for consultants.Culture 2 • Consultant doctors do not attend haemovigilance andtransfusion educational updates• No details on one caseOther 3 • Training had not been provided- on issue not solely related to transfusion,- to a single person (no training plan in place to identify thisperson had not previously attended)• Information on an antibody status of a patient not availablewhen patient transferred to another hospitalThe most frequently recurring system errors for bothclinical IBCT and SAE in the HBB are presented andcompared in IBCT/SAE Table 19.IBCT/SAE Table 19: Recurring system error inclinical IBCT and SAE in HBBsHuman ErrorOne hundred and eighty human errors werereported across 117 cases. There were more thanone human error reported in 63 cases. No humanerror was reported in five cases.Clinical IBCTSAE in the HBB1 Lack of policies and Lack of policies andproceduresprocedures2 Materials Materials3 Management Designpriority/Culture80

IBCT/SAE Table 20: Human errors leading to events (n=180)Classification N Commentof human errorFailure to adhere to policies 60 • Both clinical and laboratory staff failed to adhere to establishedand procedureshospital policies resulting in SAE /IBCT.Knowledge 47 • Both clinical and laboratory staff failed to apply knowledgeresulting in SAE /IBCT.Co-ordination /Communication 21 • Communication failure on specific patient care issues /transfusion event within and between disciplines• Failure to seek out or clarify specialist advice wherepractitioners lacked specialist knowledge on transfusionpractices.Slips 17 • This type of error occurred predominantly in the laboratory butalso in the clinical area.• At least seven events (30%) occurred outside routine workinghours and involved either laboratory staff (some of whom didnot normally work in the HBB) or doctors.• Five events occurred during an emergency and a further twoevents when clinical areas were extremely busy.• One event occurred when the clinician involved sought tocorrect an error.Carrying out task incorrectly 16 • These errors occurred both in the clinical area and in the HBB.• Examples include labelling errors in HBB, and collection andstorage errors in the clinical area.Verification 11 • Verification errors occurred both in the clinical area and in theHBB.• Examples include a failure to verify results prior to transfusion(clinical errors) and failure to verify unit numbers prior to issue(HBB).Monitoring 4 • All clinical errors.• Failure to monitor administration of transfusion.• Failure to monitor outcome of transfusion.Patient related 1 • Post natal patient caring for new born twins unable to complywith restrictions on arm movements, thereby prolongingtransfusion time over six hours.Qualifications 1 • This error was made by a junior doctor on a team, who did nothave a specialist qualification or knowledge of patient caseload.Unclassifiable 1 • This case has been described - IBCT/SAE Case History 21.Other 1 • Details unclear in this case – knowledge deficit suspected in anunnecessary transfusion where a patient with iron deficiencyanaemia received a red cell transfusion.81

The most frequently recurring human errors for bothclinical IBCT and SAE in the HBB are presented andcompared in IBCT/SAE Table 21.IBCT/SAE Table 21: Recurring human error inclinical IBCT and SAE in HBBs82Clinical IBCTSAE in the HBB1 Knowledge Failure to adhereto policies andprocedures2 Failure to adhere topolicies and procedures3 Co-ordination / SlipCommunicationKnowledgeThis analysis of error cause allows identification ofcontributing factors common to most events. Similarto findings in 2008, human failures – failure toadhere to policies/procedures and lack ofknowledge and system failures – lack ofpolicies/procedures governing processes and designwere the most frequently reported causes of events.Key Points• Adverse event review and reporting is avery powerful way of organised learning inorganisations in general and also intransfusion services. The information gainedfrom the identification and analysis ofadverse events will enable the identificationof gaps in the transfusion service andperhaps other services in the hospital whichrequire attention. This data can be used toidentify trends and patterns of events whichreoccur and have potential to cause harm topatients, and facilitate development ofappropriate strategies to enhance patientsafety (Commission on Patient Safety andQuality Assurance, 2008).• Use of a formal protocol will ensure asystematic, comprehensive, and efficientinvestigation, and will minimise thepotential of simplistic explanations androutine assignment of blame (Vincent et al,2008).• The NHO has worked with the clinical riskadvisors in the Clinical Indemnity Scheme(CIS) to ensure all haemovigilance staffreceive system analysis root cause analysistraining. Furthermore the implementation ofthe recommendations of the Patient SafetyCommission will include a national roll outof an agreed approach to systems analysisto all health care organisations.A review of reports received in 2009 indicated 56hospitals reported 60 corrective actions and /orfurther review of practices, 51 hospitals reported noaction following an adverse event and informationwas unavailable in 15 reports.IBCT/SAE Table 22: Follow-up action forSAE/IBCT ( n=60)Category offollow-up Details N 10actionProcess, IT andequipmentchangesEducation andTraining• Change of current work process• Addition of alert sticker• Addition of an IT warning• Order new equipment 14• Includes both generalhaemovigilance training andtargeted updates.• Targeted all staff• Included an expansion ofcompetency assessment 21Communication • Memos to staff• List of implicated patients tobe sent to HBB• Direct follow-up with staffinvolved in SAE/IBCT 13Development • Development of new policiesand revision of • Revision /change to currentpolicies policies 8Audit andresearchOtherdepartment• Audit of current clinical practice• Area of SAE /IBCT becameresearch topic of an MSc student 2• Report referred to the clinical risk• Movement of stock i.e. transfusionadministration sets separatedfrom fluid administration sets.• Contact made with supplier repossibility of a name change 210Some reported SAE /IBCT had several follow-up actions includingboth corrective and preventative measures.

Key Points• The NHO recommends that change shouldbe introduced following a systematicreview of the event. Introduction of achange should include development ofpolicies to support practice change,informing all relevant stake holders andprovision of training to ensure that theinformation on change is bothdisseminated and acted on.• Haemovigilance and transfusion servicesshould then engage in monitoring not onlyto evaluate the impact of the change interms of transfusion service, but thepotential to impact on other hospitalservices. This follow-up monitoring iscrucial to ensure ongoing learning andimprovement, and is characteristic of aquality service.83

Paediatric Adverse Events2009Paediatric cases comprised 20% of all SAE/IBCTcases in 2009, with 25 reports accepted. Thirteencases (52%) were mandatory SAEs and 12 (48%)cases were non-mandatory IBCTs. Two casesinvolved anti-D Immunoglobulin (Ig) and aredescribed on page 91, and one case is describedin the factor concentrates section on page 89. Ofthe remaining 22 cases the majority (n=17, 76%)involved red cells (IBTS/SAE Figure 9).Findings.IBTS/SAE Figure 9. Components associatedwith Paediatric SAE/IBCT (n=22)3 (14%)1 (5%) 1 (5%)Red CellsPlateletsSD PlasmaFFP17 (76%)The categories and age groups of SAE/IBCT are summarised in IBTS/SAE Table 23.IBCT/SAE Table 23: Paediatric SAE/IBCT in Table 1: (n=22).Category n Major Moderate Neonate Infant Infant Child

Transfusion of the Incorrect Component/ Product(no reaction) (n=11)Eleven cases were captured in this category makingup half (50%) of all paediatric cases.Increased donor exposure due to problems withpaedipacks (n=6)• In the first case, red cells suitable for neonataluse were transfused instead of a paedipack. Thisoccurred when a unit of red cells was issued to asatellite fridge for an infant who required asurgical procedure. The infant later required atop-up transfusion, but instead of requesting apaedipack from the transfusion laboratory, thered cells in the satellite fridge were transfused.• In two further cases, infants were exposed toother donors, even though aliquots from aprevious paedipack were still available. In thefirst case, an infant was transfused with twoaliquots from a paedipack. The baby was thendischarged and, subsequently readmitted. Whenthe transfusion laboratory received a request foranother transfusion, a second paedipack wasordered from the supply centre. The error wasattributed to a lack of communication as theblood bank staff did not realise that aliquotsfrom the first paedipack were still available.In the second case, two errors occurred. Theinitial error was when a second paedipack wasordered although an aliquot from the existingpaedipack was still available in the blood bank.A further error occurred when anotherpaedipack was ordered even though threealiquots from the second paedipack were still instock. Root cause analysis attributed this error toa considerable increase in the workload of theblood bank staff.• In a further case, a neonate was exposed to anadditional donor when an aliquot of a paedipackwas issued and the remaining aliquots were notreturned to controlled storage.• Another case involved a neonate where theremaining aliquots from a paedipack wererecalled by the supply centre for suspectedbacterial contamination due to a false positivebacterial alert on an associated plateletcomponent. This was a mandatory SAE and iscaptured under ‘materials’ on page 69.• In an additional case which was categorised as amandatory SAE, an on-call medical scientistordered an incorrect component from the supplycentre instead of a paedipack for an infant.Although the error was noted and queried by amedical scientist, a more senior scientistincorrectly advised it was suitable for issue. Inthis case, the baby required no furthertransfusions.Other (n=5)• In one case a unit of red cells was issued underthe hospital neonatal policy, which permits theissue of uncrossmatched red cells. In thisinstance however, the infant was several daysolder than the age limit specified in the policy.• In two cases, pooled platelets were issuedinstead of apheresis platelets, while anothercase involved the issue of fresh frozen plasma(FFP) instead of SD plasma.• In another incident an on-call medical scientistissued universal SD plasma (Uniplas) instead ofGroup A SD plasma (Octaplas) for a group Aneonate.Transfusion of Incorrect ABO group (n=1)IBCT/SAE Case History 24This case involved a one month old B RhDpositive baby who required a transfusion of redcells. No maternal sample was available as thebaby was referred from another hospital. Inthese circumstances either a crossmatchbetween the red cells and the neonatal serummust be undertaken to exclude passive A or Bantibodies, or group O red cells must be used.In this case a B Rh D positive unit was selectedwithout a crossmatch. Root cause analysis (RCA)found the error was made by an on-call medicalscientist not normally working in transfusion. Theinfant suffered no sequelae.Transfusion of Other Antigen Incompatible RCC(n=1)IBCT/SAE Case History 25This case involved a neonate whose mother wasRhD negative with anti-D, anti-E and anti-Jk aantibodies. Following delivery, the baby whowas RhD positive developed haemolytic disease85

of the newborn with associated jaundice and anelevated bilirubin. The patient was transferredto a neonatal unit in another hospital, andrequired a transfusion for a Hb of 7.9 g/dl. AnO RhD negative paedipack was ordered fromthe supply centre, but no patient details weregiven, nor was the supply centre informed ofthe requirement for a Jk a negative component.The baby was transfused with a Jk a positiveunit. On the day prior to the transfusion, theHBB had received an antibody report on thematernal blood sample from the supply centreindicating the previous antibody history. Noconnection, however, was made between thematernal sample and the unit for transfusion tothe baby, who suffered no sequelae.Unnecessary Transfusions (n=2)Two patients received unnecessary transfusionsbased on incorrect haematology results.• In the first case, a neonate in a critical conditionin ICU was given an urgent transfusion of redcells. An FBC checked in the haematologylaboratory taken earlier in the day recorded thebaby’s Hb as 13.5 g/dl. However, later thatnight a sample was checked on a blood gasanalyser which incorrectly read the patient’s Hbas 8.8g/dl. No sample was sent to thehaematology laboratory to verify this result aswas hospital policy. A clinical decision wasmade by the anaesthetist to transfuse thepatient. The error was discovered the next daywhen a medical scientist realised the Hb of 8.8g/dl on the request form did not correspondwith the Hb result recorded in the laboratory.The post transfusion Hb on the following daywas 16.2 g/dl.• In the second case, a two month old infantreceived one and possibly two unnecessarytransfusions. The infant’s Hb was recorded as14.3 g/dl prior to surgery. However, the nextday following a surgical procedure, the infanthad a cardiac arrest, and was transfused with 45mls of red cells based on a Hb of 11.2 g/dlprocessed using a blood gas analyser. A samplewas taken the next day for an FBC, but it wasclotted, therefore the baby was transfused witha further 45 mls based on results again usingthe blood gas analyser which read the Hb as17.7 g/dl. Because repeated FBC samples wereclotted and could not be tested, no further Hbresults were available until six days later, whenthe repeat Hb was 19.4 g/dl.Transfusion of Incorrectly Stored Component(n= 1)There was one report of transfusion of acomponent that was out of controlled storagebeyond the recommended time. The case involveda neonate whose condition deteriorated after theunit was removed from the fridge. Thisnecessitated a procedure which delayed thecommencement of the transfusion.Transfusion of Incorrectly Labelled Component(n=1)IBCT/SAE Case History 26.An aliquot of a paedipack was issued for aneonate in one name. However, there was achange in the baby’s name and a second pretransfusionsample was not requested or sent tothe laboratory, as per policy. When a secondaliquot of the paedipack was requested, thelaboratory issued it using the original namelisted on the request form. The error was furthercompounded as the patient did not have anidentity band, although two identity bands wereattached to the patient’s cot. The error wentundetected during the pre-transfusion bedsidecheck and was only discovered duringretrospective checking of ‘on-call’ work.Other (n= 3)In three cases, there was a failure to use the correctfilter. In one case, a blood filter was not used totransfuse a unit of red cells. In the second case, ablood filter was attached to a unit of red cells, butwas by-passed and the component administeredvia a syringe. In the final case, a rare unit of HPA 1anegative platelets was wasted due to the selectionof the incorrect filter. At the time there were nohospital guidelines to indicate the correct filter tobe used.Failure to give CMV negative &/or irradiatedcomponent (n=2)There were two cases in this category. In one case,the doctor failed to request a CMVnegative/irradiated component for an infant withspecial requirements. In the second case, theclinician requested irradiated red cells for an infantbut, the medical scientist issued non-irradiated red86

cells in error. This patient may not have in factrequired irradiated red cells and neither patientsuffered sequelae.Root cause analysisThe majority of paediatric cases (n=13, 59%) wereclassified as ‘major’, that is, having a high risk ofcausing harm to the patient. The findings of RCA aspresented in IBTS/SAE (Table 24) found all casesinvolved human error and system errors were foundin four cases. In some cases more than one erroroccurred.IBCT/SAE Table 24 : Root Cause Analysis ofPaediatric IBCT/SAE (n=40)IBTS/SAE Table 25: Where Paediatric errorsoccurred and who was involved (n=24)Practitioner n =24(%) Department n =24(%)Nurse/Midwife 5 (21) Laboratory 12 (50)Doctor 6 (25) Ward 3 (13)MedicalNeonatalScientist 13 (54) Unit/ICU 6 (25)Supply centre 1 (4)Human Failure n System Failure nFailure to adhere topolicies &procedures 10 Management Priorities 1Theatre 2 (8)Total 24 24Knowledge 10 Policies/ Procedures 2Co-ordination/Communication 5 Materials 1Verification 2 Design 1Monitoring 1 Other 1Slip 5Carrying out taskincorrectly 1Total 34 6Who was involved in Paediatric IBCT/SAE andwhere did they occur?Further analysis of Paediatric SAE/IBCT as shown inIBTS/SAE (Table 25) found half (50%) occurred in thelaboratory. In some cases, more than oneperson/department was involved in the error.Key Points• The actual number of paediatric cases from2008 to 2009 remains almost unchanged,although the percentage of reports havedecreased slightly from 22% in 2008 to 20%in 2009.• Of concern, however, is that analysis ofIBCT/ SAE in terms of potential to causeharm to patients showed that 59% (n=13) ofpaediatric cases had high potential to causeharm, compared with 36% (n=36) of reportsin the adult population. This highlights therisks to paediatric patients receivingtransfusions. Other haemovigilanceschemes have reported similar findings(Stainsby, 2008).• Of further concern is that 86% (n=19) ofpaediatric cases in 2009 involved neonatesand infants.• Further analysis also found that mandatorySAEs comprised more than half (n=13, 59%)of all paediatric cases compared to only33% (n=33) of all adult cases.87

• Several patients were unnecessarilyexposed to another donor, despite thefact that aliquots from a previouspaedipack were still available.Recommendations• As highlighted in previous NHO reports,paediatric patients have specialised bloodrequirements and errors may have serioussequelae. The high incidence of errorscategorised as ‘high risk’ and involvingneonates and infants suggest the need forparticular vigilance in this area.• Where antibody reports are received from thereference centre indicating the presence ofantibodies, HBB staff should check theseagainst the patient’s current transfusion needs,or in the case of an infant, the mother’santibody history, to ensure antigen negativeblood is provided.• Wherever possible, blood components shouldbe prescribed based on results from a sampleanalysed in the haematology laboratory, andnot on results from gas analysers.• It is essential that hospitals have policies andguidelines on best practice, particularly forinfants under four months as detailed in theguidelines Transfusion of Blood Components toInfants under Four Months published byNational Blood Users Group (2007).• Laboratories providing blood components forpaediatric patients must be adequately staffedwith well-trained personnel (UK TransfusionCollaborative, 2009).• The BCSH (2009), guidelines on theadministration of blood components stipulatethat clinical staff involved transfusion practicereceive regular education and training. This isparticularly important where staff and skills maynot be utilised on a daily basis. Variousmethods are suggested such as face-to-face,self-directed learning and e-learning. Thepaediatric module of the e-leaning programme(http://www.learnbloodtransfusion.org.uk) inblood transfusion practice is a usefulsupplement for ongoing educationrequirements.• There should be procedures in place to ensureall staff especially those on-call are aware ifaliquots from a paedipack are still available inthe HBB. Where aliquots are removed fromcontrolled storage for issue, the remainingaliquots must be returned to the fridgeimmediately.88

IBCT Involving FactorConcentrates/Blood Products2009Errors surrounding coagulation factorconcentrate (n=8)Key Findings• Three adverse events occurred inresponse to an emergency bleed.• A majority of adverse events (71%)involved failure to follow specific hospitalor checking policies.IBCT/SAE Table 26: Reports relating tocoagulation factor concentrates (n=8)Nature of Report n Site of ErrorIncorrect patientreceived product 1 AdministrationIncorrect productadministered 4 Prescription/RequestHBB errorDistributionIncorrect dose ofproduct administered 3 Administration (all)Incorrect patient received product (n=1)• This occurred late at night, when a doctor, whohad been working throughout the day, failed tocarry out patient identification checks andadministered coagulation factor concentrate(CFC) to the wrong patient.Incorrect product administered (n=4)incorrect product. Neither the doctor nor theHBB contacted the on-call haematology teamfor verification.• The surgical team looking after a patient withliver dysfunction sought specialist advice from ahaematology team in another hospital fortreatment of bleeding and mild coagulopathy.The advice from the haematology team was totreat with SD plasma and platelets. However,the surgical team were confused and requestedtwo vials of prothrombin complex (PCC) in errorfrom the HBB. The patient received PCC (whichshould only be used in liver disease onspecialist advice) without any adverse outcome.Both doctors involved in this event wereconfused by the similar names for PCC-“Octaplex“ and SD plasma “Octaplas“.• In the second case, a medical scientist on-callissued plasma derived CFC in place ofrecombinant product which had been ordered.• In the final case, the patient required anemergency delivery of CFC to his home. Thepharmacist on-call did not complete all checksprior to distribution and the patient receivedthe incorrect product.Key Point• The similarity in the trade name for bothSD plasma and PCC has caused confusionespecially for clinical staff. In this year’sreport, the use of this resulted in thewrong product being issued to thepatient.- Products should be ordered as SDplasma and PCC, rather than Octaplasand Octaplex, to avoid this type ofevent.- The NHO has made this concernknown to both the manufacturer andthe IMB.Two adverse events occurred at prescriptionrequest.• A doctor in the ED ordered the incorrectproduct for a patient with inhibitors despiteadvice from the HBB. The HBB then issued the89

Incorrect dose of product administered (n=3)• All adverse events occurred at administration.Two cases involved patients receiving a largerdose of CFC than required.• The final case involved administration of PCCfor warfarin reversal. The patient received lessproduct than prescribed following an errormade by a nurse when setting up the infusionpump. The pump was set to deliver 3mls/hourwhen it should have delivered 3mls/minute,resulting in significant under-dosage to thepatient. Medications are usually administered inmls/hr, but clinical staff - especially nursesshouldpay particular attention to medicationorders.Key Recommendations• Management of haemophilia patients withinhibitors is complex and expert coagulationadvice should be sought.• Medications are usually administered in mls/hr,but clinical staff especially nurses should payparticular attention to delivery of medication.• All cases of incorrect dose of product involvedcalculating calculating correct dosage and rateof administration of product. Second checksmust be independent of the first check tominimise the potential for passive checking byone or both persons involved in the check.90

Adverse Events associated withAnti-D Immunoglobulin 2009IntroductionTwenty seven reports were received in 2009.Seventeen reports related to delays in anti-D Igadministration, eight were unnecessaryadministration of anti-D Ig and two were omissionof anti-D Ig.Anti-D Figure 2 : Reports received in 2009(n=27)Anti-D Table 3: Patients affected by Anti-D IgIBCT (n=27)Ante-natal Post -natalDelay in administeringAnti-D Ig 14 3Omission of Anti-D Ig 2 0Unnecessaryadministrationof anti-D, 8Ommisions of anti-D, 2Unnecessaryadministration ofAnti-D Ig 4 4Delays inadministeringanti-D, 17Total 20 7Reporting establishmentsTwenty hospitals in Ireland offer maternity servicesto patients. Reports were received from 10hospitals in 2009. All dedicated maternityhospitals submitted reports. No reporting patternwas identified based on hospital activity. Howeveran active haemovigilance programme possiblyincreased reporting. A lack of awareness ofreporting may exist especially in general hospitalswith maternity units.Key Point• Haemovigilance Officers working with thehospital transfusion teams should examinehospital systems for identifying andreporting adverse events relating to anti-Dto ensure these are effective.Patients affectedTwenty seven patients were involved in reports in2009 two patients were under 18 years of age.Twenty (74%) of these SAE/IBCT occurred inantenatal patients in 2009.Delay in administering Anti-D Ig (n=17)There were 17 reports of delay in administration ofanti-D Ig in 2009. The majority of these reports (14)related to ante-natal patients, and three to postnatal patients. Thirteen delays involving ante-natalpatients occurred in either a Day Ward orEmergency Department (ED) setting.There were at least five cases, where reportsindicated patients were discharged prior to eitherresults being available or anti-D Ig issued from theHBB. One case is described.Anti-D Case History 8A 29 year old female patient, 24 weeksgestation, presented to the ED on Sundaymorning, approximately 18 hours following a fallover a bank holiday weekend. A sample wastaken for baseline blood group and antibodyscreen. The HBB did not process this sampleuntil the following Tuesday morning. Therequest form did not include any details as towhen the fall had occurred and anti-D Ig wasissued later that evening which was greater than72 hours following the sensitising event. The91

patient was unable to attend the ED and onlyreceived anti-D Ig five days after the sensitisingevent.There was no information available on whetherthese patients became sensitised following thedelay in anti-D Ig administration.Unnecessary Administration of anti-D Ig (n=8)Eight cases of unnecessary administration of anti-Dwere reported in 2009. Four antenatal as well asfour post-natal patients were affected. Thesereports were classified as follows.Anti-D Table 4: Breakdown of unnecessaryadministration of Anti-D.Administered to a Rh D 1Positive WomanAdministered to mother of 1Rh D Negative babyAdministered to a previously 5immunised patientAdministered based on 1expired prescriptionAdministered to a Rh D positive female (n=1)In this case, initial testing of the patient sample onan automated analyser suggested the patient wasRh D negative. Further manual testing showed thepatient to be Rh D positive. This led tounnecessary administration of anti-D Ig to an antenatalpatient.Administered to mother of Rh D negative baby(n=1)92Anti-D Case History 9In this case, a postnatal patient received anti-DIg unnecessarily. Both the mother and babywere Rh D negative but anti-D Ig wasprescribed in the patient’s transfusion record.The reason for this prescription was neverclarified. Anti-D Ig was issued by the HBB tothe clinical area on the basis of the order. Therewas no policy in the HBB to check the patient’sblood group prior to issuing anti-D Ig.nThe nurses working administered the anti-D Igto the patient. They assumed that when it hadbeen issued from the HBB, the patient shouldreceive it. The postnatal ward was covered bylocum staff nurses from the gynaecology ward,where it would not be routine practice toadminister anti-D Ig.Since this event, the HBB staff must now checkthe patient’s blood group prior to issuing anti-D Ig.Administered to a previously immunised patient(n=5)Three cases involved postnatal patients and twoantenatal patients who were already sensitised anddid not require anti-D Ig.Post natal patients• A patient known to have immune anti-Ddelivered a Rh D positive baby. Anti-Dquantitation had been carried out about 8 daysprior to delivery. Following delivery, theobstetric registrar on call contacted thehaematology team and was advised not toadminister anti-D Ig. However, the obstetricregistrar subsequently prescribed anti-D Ig. Thebasis for the clinical decision remained unclear.• In a second case, a Rh D negative patient haddelivered a Rh D positive baby. Results of thefinal antibody screen were not filed in thepatient’s notes. While the final screen waspositive, results from previous screening werenegative. Anti-D Ig was administered on basisof results available in the clinical notes.• In the third case, the patient’s antibody statuswas not verified prior to either prescription oradministration of anti-D Ig.In the last two cases, a stock of anti-D Ig wasmaintained in the clinical area and therefore wasnot under control of the HBB.Antenatal patientsThese two events were caused by knowledgedeficits of medical staff.• In the first case, the doctor caring for a patientwho had an antepartum haemorrhage did notwait for the result of the antibody screeningtest.• In the second case, the doctor was aware thatthe patient had formed anti-D, but mistakenlythought administering anti-D Ig would preventa further boosting of the level following a PVbleed.

Administered based on expired prescription(n=1)This antenatal patient attended with a urinary tractinfection. She was administered Anti-D Igunnecessarily on basis of an expired prescription,relating to a previous admission for a sensitisingevent.Omission of Anti-D (n=2)There were two cases, both involving antenatalpatients. Neither patient was reported as havingbecome sensitised.Anti-D Case History 10A patient at 16 weeks gestation was having anamniocentesis. The clinical team received anincorrect verbal report from the HBB that thepatient’s blood group was Rh D positive andthe patient did not receive anti-D Ig. Followingthis event, the HBB no longer provides clinicalstaff with a verbal report on the patient’s bloodgroup, and the clinical staff must now check iton the LIS.Anti-D Case History 11A patient at 14 weeks gestation attended theED following a antenatal sensitising event.Although she was reviewed by a doctor andher bloods were taken for blood grouping,there was no follow–up of the results. Theomission was subsequently discovered at thepatient’s (initial) planned booking visit to thesame hospital, where it was noted the patientwas RhD negative.Following this event, a formalised follow-upprocedure for the review of results andadministration of anti-D Ig has been introducedin the ED.Where did the error occurred & who wasinvolved?A review of the 2009 reports revealed that themajority of SAE /IBCT 17 (63%) occurred in theclinical area. Doctors were involved in eight (47%)and midwives in three (18%) clinical SAE /IBCT.While it was evident a clinical health careprofessional was involved in the six remaining(35%) events, it was not clear whether it was adoctor or a midwife.11Site of error for one AE was both the clinical area and HBB.Three errors in the HBB resulted in patientsexperiencing a delay, an omission and anunnecessary administration of anti-D Ig.The site of error in seven cases was classified asOther ( Anti-D Table 5).Anti-D Table 5: Site of error (n=20)HBB Clinical OtherDelay inadministeringAnti-D Ig 1 10 6Omission ofAnti-D Ig 1 1 0Unnecessaryadministration 1 6 11 1Total 3 17 7Site of error was described as Other in followingcases:• Five patients were discharged prior toreceiving anti-D Ig. In all cases blood sampleshad been taken for screening- One postnatal patient was discharged fromthe labour ward. This patient had receivedanti-D Ig prior to delivery following a fall.Following discharge, the clinical nursemanager contacted the patient to returnthe following day for her postnatal dose.The patient did not return until 72 hoursafter delivery.- Four antenatal patients who attended theED for sensitising events were dischargedas samples were not processed in the HBBoutside routine working hours. In twocases, patients indicated they were unableto return at the time advised. One of thesecases is already described above in anti-DCase History 8.• One patient had opted for shared care andattended her General Practitioner (GP)following a fall. The GP did not consider the fallto be a sensitising event and did notadminister anti-D Ig. This patient received anti-93

D Ig during a follow –up visit to her ante-natalclinic in her hospital.• In the final case, an already immunised patientreceived an unnecessary dose of anti-D Ig. Thisoccurred because the clinical staff wereunaware of the more recent results, as thereport had not been filed in the chart(previously described on page 92).Cause of errorSince 2007, the NHO has used root cause codesfrom Medical Event Reporting System-TransfusionMedicine (MERS-TM) to classify causes ofIBCT/SAE.A review of reports revealed multiple causes withboth human and system failures contributing toIBCT/SAE.Key PointThere were five cases where the patient wasimplicated in the error. These cases involvedpatients who were discharged prior toavailability of screening results, and thesepatients did not / were unable to return.Surprisingly the reporting hospital did notreport the delay in sample processing orfailure to process samples outside routinehours as a system failure.The following human errors were reported.Anti-D Table 6 : Human error contributing toIBCT/SAE (n=35)Human ErrorFailure to adhere to policies/procedures 15Knowledge 10Patient related 5Carrying out task incorrectly 4Verification 1• Failure to adhere to hospital policies andknowledge deficits were the most frequentlyreported human errors. Clinical staff weremainly implicated.- Clinical staff (doctors n=5; midwives n=3,clinical healthcare professional n=5) wereimplicated in 13 of 15 cases where failure toadhere to hospital policies was indicated asa cause of error.- Where knowledge deficit was identified as acause of error medical staff were implicatedin six cases, midwifes in three cases, andclinical healthcare professional (doctor,midwife or nurse) in three cases.N94Anti-D Table 7: System error contributing to IBCT/SAE (n=7)System Error N CommentSystem errors were reported in seven cases asfollows.Lack of policy 2 • The policy on prescribing anti-D was unclear (Unnecessary Administration).• There was no defined responsibility for collecting samples for blood groupin theatre. In an emergency case, bloods were not taken (Delay in Administration).Management 2 • Staff shortages in postnatal ward, resulted in inexperienced staff (from aPrioritygynaecological speciality) working in this ward (Unnecessary Administration).• HBB is not staffed to process samples out-side routine working hours,resulting in a delay in processing of samples (Delay in Administration).Other 2 • In one case, a doctor did not take a sample for blood group, when an ante-natalpatient presented following a sensitising event. This was due to a lack ofknowledge. The doctor had not received training. Medical training is provided atinduction and is unavailable outside these times (Omission of anti-D Ig).• In the second case, results from the HBB were not filed in the patient’s chart. In thisorganisation, anti-D Ig was not issued from the HBB, but was available from theclinical area. (Unnecessary Administration).Design / 1 • In this case, an error in blood grouping resulted in an antenatal patientEquipmentreceiving anti-D Ig unnecessarily.

Silent SensitizationsIt is estimated that 1% of Rh D negative womendevelop anti-D antibodies due to small or silentbleeds occurring in the final trimester of pregnancy(Mollison et al 2005). Postnatal anti-D Igadministration is too late for these women. Theefficacy of routine antenatal prophylaxis has beenshown in many studies (Tovey et al, 1983; Lee andRawlinson, 1995) and has been accepted by theNational Institute for Health and ClinicalExcellence (NICE) (2008). However, routine antenatalprophylaxis remains unavailable in Ireland.The NHO does not collect incidences of silentsensitisations, but will continue to collect reportsinvolving cases where sensitisations occur or havepotential to occur following an adverse eventespecially if it relates to delay or omission inadministration of anti-D Ig.Recommendations• Midwifery, medical and blood transfusionlaboratory staff need to be fully familiar withcurrent best practice surrounding appropriateand timely administration of anti-D Ig. Clearprotocols and criteria should be in place foranti-D-Ig administration and assumptionsshould not be made that because anti-D Ig hasbeen issued by a HBB that the patient shouldreceive it. This has implications for professionalbodies, universities and hospital trainingdepartments including haemovigilance.• Follow –up of potentially sensitised patientsshould be carried out locally.- The NHO recommends that hospitalsshould develop a system to monitor andrecord incidences and outcomes of silentsensitisations.• Consideration should also be given to thepossibility of establishing of nationalsurveillance system for these events.• Recommendations from 2008 still apply.• Of concern is the increasing number of reportsof delay and omission of anti-D Igadministration to women in the antenatalsetting. Anti-D Ig should be administered assoon as possible after a sensitisation andalways within 72 hours after an event (BCSH,2006a). Hospitals should review both clinicalwork processes and the prioritisation oflaboratory testing for patients with potentiallysensitising events. Where patients withpotential sensitisation attend the emergencydepartment /obstetric unit outside routinelaboratory working hours and especially overlong weekends or holiday times, it is importantthat robust procedures are in place to ensurethat the appropriate samples are taken,analysed and results acted upon to ensure thatanti-D Ig is administered within therecommended time frame. If patients aredischarged prior to availability of results thenthis must be with the reassurance that follow-upwill occur. Delay in sample taking and testingadversely impact on optimal time foradministration of anti-D Ig for all patientsespecially those who may not immediatelyattend the ED/ obstetric unit.95

Serious Adverse Reactions (SAR)2009There were 110 SAR. There were no reports ofTransfusion Associated Graft versus Host Disease(TAvGHD), Post Transfusion Purpura (PTP) oradverse donor reactions related to predepositautologous transfusion (PAD).There was onereport received as a possible TRALI howeverfollowing extensive review this case wasreclassified as TACO.Acute Transfusion Reaction (ATR)During the reporting year 2009, 71 reports ofATR were reported. The breakdown is given inSAR Table 16 below.SAR Table 16: Acute Transfusion Reactions(n=71)Febrile Non Haemolytic Transfusion Reaction 37Acute Allergic and AnaphylacticTransfusion Reaction 28nUnclassified Reaction 3Hypotensive Reaction 1Acute haemolytic transfusion reactions 2Total 7196

Acute Haemolytic Transfusion Reaction(AHTR) (n=2) 2009FindingsTwo acute haemolytic transfusion reactions werereported in 2009. Both reactions wereimmunological haemolysis due to other alloantibody.SAR Case History 12In the first case the patient was admitted tohospital and was transfused on two occasions.There was no report of a transfusion reaction.Following this the patient was transferred toanother hospital for further management andrequired additional transfusions. Thepretransfusion antibody screen testednegative and the patient was transfused withtwo units of RCC. During the transfusion of thefirst unit, the patient developed a very slighttemperature rise (0.5˚C). The second unitcommenced and approximately two hours intothe transfusion the patient developed atemperature rise (1.2˚C), nausea, vomiting andtenderness in right iliac fossa.SAR Case History 13In the second case a paediatric patient with anunderlying congenital haemolytic anaemiarequired a transfusion for a Hb of 5.9g/dl.Three hours 45 minutes into the transfusionthe patient developed a temperature rise,rigors and haemoglobinuria.Investigations post transfusion identified anincrease in bilirubin and LDH. The units issuedwere compatible with the patient’s sample.Following referral to an international referencelaboratory (IBGRL) additional antibodies weredetected – anti-Le a + Le b . However theseantibodies are not normally associated withhaemolysis. On review, a clinical diagnosis ofHyper-haemolytic syndrome associated withcongenital haemolytic anaemia was made onthis patient. The patient received furthertransfusions since this incident with nosequelae.Later that day the patient required surgery.Intra-operative reported blood loss was800mls and a further unit was transfused.There was no report of any transfusionreaction at this time. Investigations posttransfusion identified an increase in bilirubin(42umol/L) and LDH (3698u/L) and an anti-Cwwas identified in the HBB.Follow up serological investigations carriedout at a reference laboratory identified anti-Cw and an anti-C weakly reacting in enzymeand IAT.It is likely that the reaction was due to anti-Crather than the anti-Cw as anti-Cw rarelycauses haemolysis. The reaction representedan acute haemolytic reaction on thebackground of a delayed haemolytic reactiondue to the development of anti-C antibodies(+/-Cw) subsequent to the original transfusionin the first hospital.97

Febrile Non Haemolytic TransfusionReactions (FNHTR) (n=37) 2009FindingsThere were 37 reports which fulfilled the criteriafor FNHTR in 2009. Thirty four of the patientswere adults. Three reactions occurred inpaediatric patients (page 112).SAR Table 17 Components implicated inFebrile Non Haemolytic Transfusion Reactions(n=37)ComponentnRCC 32Apheresis Platelets 1Pooled Platelets (in platelet additive solution) 3Multiple ComponentsRCC, SD Plasma, Apheresis Platelets 1Total 37Clinical outcomeThe clinical outcome was given in all cases. Thirtythree patients fully recovered, two patients hadminor sequelae with one patient requiringovernight admission and two patients diedunrelated to the transfusion.98

Acute Allergic and AnaphylacticTransfusion Reactions (AA) (n=28) 2009FindingsThere were 28 reports which fulfilled the criteriafor this type of reaction. Nineteen reactionsoccurred in adults and nine reactions occurred inpaediatric/adolescent patients.SAR Table 18 Components implicated in AcuteAllergic and Anaphylactic Transfusion Reactions(n=28)ComponentnRCC 9SD Plasma 1Apheresis Platelets 14Pooled Platelets (in platelet additive solution) 3Multiple ComponentsRCC+ Apheresis Platelets 1Total Reactions 28CommentAs in previous years FNHTR were more commonwith red cells and AA were mainly associated withplatelet components. The incidence of combinedFNHTR and AA per type and dose of platelettransfused is given below in SAR Table 19.SAR Table 19 AA and FNHTR per type of platelet component issued in 2009 (n=21)Type of platelet issued Number issued in 2009 Incident of reaction per unit issuedPlatelets pooled in PlateletAdditive Solution 8553 1 per 1426 units issuedLeucodepleted Plateletspooled in plasma 602 No reactions reported to this componentApheresis Platelets 17,173 1 per 1145 units issuedClinical OutcomeThe clinical outcome was given in all cases. Twentyfour patients made a complete recovery. The finalfour patients had minor sequelae but recoveredfollowing intervention.99

Unclassified Reactions (n=3)2009FindingsA total of 13 reports were originallysubmitted as unclassified reactions.Following review, six cases wererecategorised (1 DHTR, 2 AA and 3FNHTR), four cases did not progress.Three reactions were accepted asunclassified. All three reactionsinvolved RCC, SAR Table 20. Onereaction occurred in a neonate andthe other two reactions occurred inadults.SAR Table 20 Symptoms associated with Unclassified Reactions (n=3)Case Age & Underlying Imputability Interval Cardiac Respiratory Chills/ Other ClinicalNo. Gender Condition between Symptoms symptoms Rigors Outcomecommencingtransfusionandsymptoms1 Neonate Complex Unlikely 3 mins Bradycardia Falling O2 Decrease Required CPR(< 28 days) cardiac saturation in respiration but recovered- male condition rate,unresponsive2 Adult Abdominal Possible 4 mins Tachycardia Yes Restlessness/ Complete(51-70 surgery anxiety recoveryyears)-male3 Adult Chronic Possible 45 mins Hypertension, Dyspnoea Yes GI symptoms Complete(51-70 renal tachycardia including recoveryyears)- failure crampsfemale100

Hypotensive Reactions (n=1)2009FindingsOne case was accepted this year. This elderlypatient involved received a unit of RCC inrecovery following surgery. Approximately twentyminutes after the transfusion commenced thepatient became hypotensive. No othersymptoms were noted. The patient was treatedwith intravenous fluids and recovered withinthirty minutes.Recommendations for management ofATR 2009Recommendations for 2008 still apply and inaddition:• As noted in the 2008 report although it maybe necessary to issue least incompatibleblood for a patient, samples should be sent atthe same time to a reference laboratory forantibody identification to minimise risks oftransfusion of incompatible blood• Reaction Alerts in patient charts and/or on thehospital patient admittance system and ITsystem can be valuable in those patients witha previous AA/FNHTR reaction to ensureappropriate component selection and premedication prior to future transfusions.101

Delayed Haemolytic Transfusion Reaction(DHTR) (n=14) 2009FindingsThis year there were 14 reactions which fulfilledreporting criteria in this category, a significantincrease on previous years.All the patients affected were adults, eleven werefemale and three male. Seven patients had aprevious transfusion history, four patients hadnever been previously transfused and in threecases a transfusion history was not available.These three cases involved female patients andthe antibodies may have been due to pregnancy.Reactions occurring as a result of an errorTwo reactions occurred as a result of an error.System failures were highlighted in both caseswith human failure also identified as the cause oferror in the second case.SAR Case History 14In this case the patient had an anti-Jk b whichhad previously been entered into her recordbut the antibody information had beenremoved inadvertently some years previously.As a result of this there was no computer flagidentifying the patient’s previous antibodystatus and a Jk b positive unit was issued andtransfused as no antibody was detected onantibody screen. Approximately two weekslater, the patient had a Jk b antibody detected,positive DAT, and raised bilirubin. No LDHresult was reported. The investigationrevealed the antibody flag had beenremoved, but was unable to establish why thehistorical antibody record was altered.SAR Case History 15In the second case the patient was admittedto hospital and required a transfusion. Asample was referred to a reference laboratorywhere an anti Fy a was detected. Two antigennegative units were issued and transfused.The patient’s consultant was notified of theantibody. Following this, the patient wastransferred to another hospital for furthermanagement and required additionaltransfusions. The receiving hospital wasunaware of the patient’s previous history oftransfusion and Fy a antigen negative unitswere not selected. Approximately five dayslater, the patient had a fall in Hb, fall inhaptoglobin levels, a raised LDH and apositive DAT. Failure in communication and asystems failure were highlighted as the errorcauses in that there was no system in place toensure that relevant transfusion history andrelevant serological information were notifiedto the receiving hospital.Clinical OutcomeThe clinical outcome was given in all of the casesreported. Thirteen patients made a completerecovery and one patient died unrelated totransfusion.Recommendations for DHTR 2009The recommendations for 2008 still apply.Additionally, there should be robust systems todetect and reconcile patients’ previous historiesand transfusion records (SHOT, 2009). If a patientis transferred to another hospital, theirantibody/transfusion history should betransmitted to the receiving hospital. This issupported by the recently published draftNational Standards for Safer Better Healthcaredocument (HIQA 2010 ) which states that serviceproviders share necessary information tofacilitate the transfer or sharing of care in a timelyand appropriate manner ( Standard 3.4 Criteria3.4.2 and 3.4.3).• An alert to medical and nursing staff aboutthe presence of red cell antibodies should beentered on the patient’s chart or electronicrecord. This draws attention to possibledelays in provision of compatible blood andthe need for transfusion advice.102

SAR Table 21 Details of Delayed Haemolytic Transfusion Reactions reported (n=14)Case Age Gender Imputability Findings Antibody Category Outcome Previous Days post ReactionNo Transfusion transfusion Caused by ErrorHistory1 Adult Female Certain Elevated Anti-Jk b Group 2 Complete Not available 14 days System Failure–design.(51-70 bilirubin, recovery System failure - other previousyears) positive antibody flag removed fromDAT LIS in error2 Adult Female Likely/ Elevated Anti-Jk b , Group 2 Complete Yes 11 days(51-70 Probable bilirubin, Anti-E, recoveryyears) elevated Anti-c,LDH, Anti-CW,fall in Hb, Anti-Fy afall inhaptoglobins3 Elderly Female Likely / Elevated Anti-Fy b , Group 2 Complete Yes 6-9 days(70+) Probable bilirubin, Anti-S, recoveryelevated Anti-CwLDH, fall inHb, positiveDAT4 Adult Male Likely / Elevated Anti-E, Group 2 Complete Yes 7-12 days(51-70 Probable bilirubin, Anti-Lu a recoveryyears) elevatedLDH, fall inHb, positiveDAT103

Case Age Gender Imputability Findings Antibody Category Outcome Previous Days post ReactionNo Transfusion transfusion Caused by ErrorHistory5 Elderly Female Likely / Elevated Anti-E, Group 2 Complete No 7-8 days(70+) Probable bilirubin, Anti-c, recoveryelevated Anti Jk bLDH,fall inHb, PositiveDAT, fall inhaptoglobins6 Elderly Female Likely / Elevated Anti E, Group 2 Complete Yes 13 days(70+) Probable LDH, fall in Anti-K, recoveryHb,fall in Anti Fy ahaptoglobins,positive DAT7 Adult Male Likely / Elevated LDH, Anti Fy a Group 2 Death- Yes 2-5 days Human Failure - co-ordination(51-70 Probable fall in Hb, unrelated and communication-HBByears) DAT positive. to unaware of recentfall in transfusion transfusion in anotherhaptoglobin hospital.level System failure – Other - nosystem in place to ensurerelevant history wasnotified to receiving hospital8 Adult Female Possible Elevated LDH, Anti-Hr Group 2 Complete Yes 36 hrs(18-30 elevated recoveryyears) bilirubin9 Elderly Female Likely / Jaundice, Anti E, Group 4 Complete No 14 days(70+) Probable elevated Anti-K, recoverybilirubin, Anti-Jk a ,fall in Hb, Anti-S.positive DATandassociatedrenalimpairment104

Case Age Gender Imputability Findings Antibody Category Outcome Previous Days post ReactionNo Transfusion transfusion Caused by ErrorHistory10 Adult Female Likely / Elevated Anti-S, Group 2 Complete No 6-12 days(51-70 Probable bilirubin, Anti-Jk b recoveryyears) fall in Hb,positive DATpre and posttransfusion.No LDH11 Elderly Female Likely / Fall in Hb, Anti-K, Group 2 Complete Not available 13 days(70+) Probable positive DAT Anti-S recovery12 Elderly Male Certain Jaundice,(70+) elevated bilirubin, Anti-Jk a Group 3 Complete Yes 4-9 dayselevated LDH,fall in Hb,positive DAT13 Adult Female Likely / Elevated LDH, Anti-Fy b , Group 2 Complete Not available 19 days(51-70 Probable fall in Hb, Anti-K, recoveryyears) positive DAT Anti-f14 Adult Female Possible Elevated Anti-S Group 2 Complete No 1-2 days(31-50 bilirubin recoveryyears)105

Respiratory Complications ofTransfusion 2009 2009Transfusion Associated Circulatory Overload(TACO) 2009 (n=18)Findings:There were 18 reports of TACO, accounting for14% of serious adverse reactions received by theNHO. This number represents more than a 50%decrease compared to 2008. Fifteen cases wereassociated with red cells, two with multiplecomponents and one with apheresis platelets.This final case was originally submitted as TRALIbut following review was reclassified as TACO(SAR Case History 18 TACO). Fourteen (77%) ofthe cases were attributed as likely/probable tothe transfusion.Twelve of the patients were females and six weremales. The majority of patients (14) 70% wereaged 70 years or more. The age and gender ofthe patients implicated in these reports areoutlined in SAR Table 22.SAR Table 22 Age/gender of patientsimplicated in TACO reports 2009 (n=18)Infant Adult Adult Adult Elderly(1-4 yrs) (18-30 yrs) (31-50 yrs) (51-71 yrs) (70+)Male 0 0 1 2 3Female 1 1 1 0 9Symptoms and underlying conditionsSymptoms of overload developed 30 minutes to7 hours after transfusion with a median onset of 3hours. The most commonly reported symptomsare outlined in SAR Table 23.SAR Table 23 Most frequently occurringsymptoms in TACO (n=18)SymptomStridor/wheeze 6Tachycardia 5Hypertension 8Dyspnoea 10Falling O2 Saturation 11nSixteen patients 84% had complex underlyingmedical problems.SAR Table 24 Underlying condition of patientswho developed TACO (n=16)Underlying conditionNo of patientsCardiac 13Respiratory 7Renal 5Underlying cardiac, respiratory or renaldysfunction was reported in 16 cases. In eightcases, patients were reported as having morethan one underlying condition and in threepatients were reported as having cardiac, renaland respiratory conditions.TACO in massive haemorrhageAgain as in 2008 two patients, a young womantreated with multiple components for massiveobstetric haemorrhage and a male patient alsotreated with multiple components for a gastrointestinalbleed with no previous underlyingcondition both developed TACO.SAR Case History 16 (TACO)This young female patient who had noprevious medical history had a normaldelivery and subsequently developed a postpartum haemorrhage. She was treated withmultiple blood components and requiredsurgery. During this time she received nineunits of RCC, two units of platelets and eightunits of SD plasma. She was brought torecovery but continued to bleed and requiredfurther surgery. Intubation on the secondoccasion was difficult and it was noted thatshe had pulmonary oedema. She was treatedwith frusemide 40mgs. Following this shereceived five more units of RCC and six unitsof SD plasma with out any further reactionnoted.106

SAR Case History 17 (TACO)In the second case a male patient wasadmitted to hospital with a gastrointestinalbleed and brought to theatre. He had noprevious cardiac, renal or respiratorycondition. In this time period he receivedseven units of red cells (2000mls approx), fourunits of SD plasma (800mls) and five and ahalf litres of IV fluids (5500mls) in total. Priorto extubation in theatre his O2 saturation fellto 86%. His chest x-ray showed residualpleural effusion on the left side He wastreated with frusemide 20mgs with goodeffect.TACO with Single Unit TransfusionsSingle unit transfusions can also result in TACOand therefore should be monitored as closely asmultiple unit transfusions (Andrzejewski andPopovsky, 2005). Ten patients developed TACOafter a single unit transfusion. The patient’s weightwas available in only five cases and complete fluidbalance information was not available in any ofthe patients. Nine patients had received diureticspost transfusion. Seven were on regular diureticsbut only three received diuretics pre/or during thetransfusion. Of these ten cases, four patients hadnot received any other components in theprevious twenty four hours. Two of these fourpatients were under 70 years and had anunderlying medical condition such as renalimpairment, cardiac failure and respiratorydisease. All four patients were on regular diureticsbut only one received diuretics pre transfusionand two had their diuretics held prior to thetransfusion.The volume of an individual red cell unit issued bythe IBTS is between 230mls to 350mls with amean of 260mls. The volume of the unitstransfused to these four patients was between276mls to 360mls each.of diuresis in the days preceding thetransfusion, the patient remained in acumulative positive balance. She hadreceived a dose of frusemide at midnight andagain at 10.00am on the day of thetransfusion. She was noted to be oedematousbut was in a negative 24 hour fluid balance of300 mls. The patient had been extubated thatday.There was a slight increase in the respiratoryrate (RR) during the transfusion and shortlyafter the transfusion was finished, but thisreturned to normal. The platelets were givenover one hour and twenty minutes in theevening and IV frusemide was given in theearly hours of the morning post thetransfusion. The exact time was notdocumented. Six hours after the transfusionwas complete, the RR increased again andwhen reviewed on rounds the followingmorning, the patient had mild stridor andincreased O2 requirements. A further dose ofIV frusemide was given. No chest x-ray wastaken on the day of the transfusion but onetaken on the following day suggestedpresence of pulmonary oedema. Following afurther x-ray the next day, a subsequentreview suggested the changes were morelikely to be due to infection.Investigations and conclusionsBecause of the possibility of TRALI the femaledonor was investigated but no antibodies to HLAI or II, antigens or granulocytes were found. Theoverall clinical findings in this case were consistentwith TACO rather than TRALI.Reactions occurring in patients as a result of anerrorTACO was reported to have occurred following anerror in two (10%) cases. Human failure was citedas cause of error and included knowledge deficit,failure to adhere to policies/procedures andfailure to monitor the patient’s Hb between units.TACO with apheresis platelets.SAR Case History 18 (TACO):The patient, a young child requiringhaemofiltration and dialysis received 150mlsof apheresis platelets for a low platelet count.The patient had been in a positive balancefrom eleven days to five days preceding thetransfusion and, although there was evidenceDiscussionConsideration should be given by the IBTS toissuing units with a standard volume and Hbcontent or making packs with aliquots availablesimilar to paedipacks. HBB should pick smallerunits for susceptible elderly patient but this canbe difficult in practice.107

Key PointAt first patients should be given a diureticprior to transfusion particularly those onregular diuretic therapy.Recommendations• Recommendations for 2008 still apply and inaddition• Doctors and nurses across all specialitiesshould receive education aimed at therecognition and avoidance of TACO. Inaddition junior doctors should receive specifictraining in the area of transfusion medicine toensure safe and appropriate decision makingregarding transfusion and prescription ofblood components/products.• As mentioned in 2008 (page 56) it is importantthat clinicians recognise that even healthypatients can develop circulatory overload inthe massive transfusion setting and that fluidbalance is carefully monitored to avoid overhydration/overloadwith components.108

Transfusion Associated Dyspnoea (TAD)2009 n=3 2009There were three cases of Transfusion AssociatedDyspnoea (TAD) in 2009. Two cases wereassociated with red cells and one with pooledplatelets. One patient suffered serious sequelaeand required ventilation for five days (SAR CaseHistory 19). Two patients had underlyingrespiratory conditions one, a pre-term infant alsohad a congenital cardiac defect and the remainingpatient had renal impairment. The mostcommonly reported symptoms were dyspnoea,hypertension, tachycardia and falling O2saturations.respiratory complications and an increase insystolic blood pressure possibly associatedwith the transfusion were suggestive of TACO.However, it did not meet the strict criteria ofthe ISBT (2006) definition of TACO, as therewere no x-ray changes or evidence of positivefluid balance, this case was re-categorised asTAD. The patient made a full recovery.SAR Case History 19 (TAD)The patient, a 31 year old male with complexmedical problems and a previous history ofrespiratory failure and an active pulmonaryinfection, received two units of red cells onconsecutive days. Approximately two hoursinto the second transfusion, the patientdeveloped tachycardia, hypertension, cyanosisand decreased O2 saturations reducing from91% on four litres of O2 to 75% on eight litresof O2 and then to 45% on ten litres of O2. Thepatient was transferred to ICU for immediateventilation to maintain airway and wasadministered antihistamine, steroids,nebulisers, anti-pyretics and frusemide. Thepatient remained ventilated for five days afterwhich he made a full recovery.SAR Case History 20 (TAD)A pre term neonate with a congenital cardiaccondition who received red cells for a low Hb,developed a temperature rise one hour intothe transfusion, and an hour later was noted tohave rapid respirations and reduced O2saturations. Blood cultures taken on the babyafter the transfusion were positive, but a fullblood count taken with the culture showed alow white cell count and neutropenia. Onreview this was considered to be consistentwith infection preceding the bloodtransfusion. Initially this case was categorisedas a FNHTR but the development of109

Suspected Transfusion TransmittedInfection (STTI) 2009 (n=4) 2009FindingsThis year four cases of suspected transfusiontransmitted infection (STTI) were reported, threecases of suspected viral infection and one case ofsuspected bacterial infection.The case of suspected bacterial infection was as aresult of a report from the IBTS associated with aconfirmed positive bacterial culture screen (BacTAlert) in a component transfused to the patient.As a precaution the patient started antibioticsfollowing the recall.Suspected Viral InfectionsThere were three viral infections reported, onereport of hepatitis B and one of hepatitis C. Inboth cases, the donor investigations werenegative, there were other risk factors andtransfusion was outruled as the possible source.The third case involved a case of possiblehepatitis A (HAV) in a haematology patient whohad a considerable number of transfusions andwho was found, on routine screening to have HAVIgM and IgG antibodies. The patient wasasymptomatic and had no evidence of abnormalLFTs. Further investigations showed that this resultrepresented a false positive and transfusionassociated HAV was excluded.110SAR Table 25 Suspected Transfusion Transmitted Infection Viral 2009 (n=3)Case Serious Age Gender Transfusion Components Donors Comments ImputabilityNo. Adverse Date ImplicatedReaction1 Transfusion Elderly Female 17-Jun-09 RCC 13 False Excludedtransmitted (70+) positiveviral infection-result.Other (HAV)2 Transfusion Adult Male 29-Mar-03 RCC, 17 15 donors Excludedtransmitted (31-50 SD Plasma, returned andviral infection years) Pooled retested HBV(HBV) Platelets negative.BloodProducts2 donors didnot return buttested negativeat the time ofdonation –Other riskfactors wereidentified.3 Transfusion Adult Female 21-Jan-00 RCC 13 All 13 donors Excludedtransmitted (51 - 70 returned andviral infection years) retested HCV(HCV) negative –Other riskfactorsidentified.

Bacterial InfectionsOne case of bacterial infection was reported. Thepatient who was transfused as a day case receiveda unit of pooled platelets for an underlyingmalignancy. No adverse reaction was noted at thetime of transfusion. Three days later there was arecall on the unit due to a positive bacterialculture alert subsequently determined to be dueto a propionibacterium acnes. As the unit hadalready been transfused, the patient wascontacted. The patient was afebrile and wasadvised to return the following day for review(four days after the transfusion of pooledplatelets). The patient’s overall condition haddeteriorated and required admission. Antibiotictherapy was commenced, blood cultures on thepatient were negative and suspected transfusiontransmitted infection was considered unlikely.SAR Table 26 Suspected Transfusion Transmitted Infection Bacterial 2009 (n=1)Case Age Gender Component Implicated Outcome ImputabilityNo.Organism1 Adult Female Platelets Propionibacterium Confirmed Unlikely(31-50 Pooled acnes positive BacT Alert.years)Patient had noreaction but wascommenced onantibiotic therapy.111

Paediatric Serious Adverse Reactions 2009“Haemovigilance systems should aim atidentifying specifically the number and types ofadverse transfusion reactions that occur in thepaediatric population because the incidence ofthe events may differ from the adult population”(Gauvin et al 2006)FindingsThis year 16 (15%) out of a total of 110 reactionsoccurred in paediatric patients. Two reactionsoccurred in neonates (

Paediatric SAR Table 5 Breakdown of paediatric reactions by component and age (n=16)ComponentsAgeSerious Adverse Red Cell Platelet Platelet Neonate Infant Infant Child AdolescentReaction Apheresis Pooled (

AcknowledgementsThe NHO appreciates the contribution of anumber of people to the compilation of thisreport. We gratefully acknowledge:The members of the NHO Team involved in thereport content compilation and writing,particularly Ms. Marina Cronin for the IBCT/SAE,Ms. Jackie Sweeney for Anti-D and FactorConcentrates, Ms. Roisin Brady for work on SAR,TRALI and STTI Ms. Kathleen Heery for TACOand Ms. Cathy Scuffil in formatting thedocument in preparation for final printing.Dr Joan Fitzgerald for her assistance with theAnti-D section and her contribution to thegeneral report contentDr Joan O’Riordan for her expert advice,especially on aspects of the STTI chapter. Heradvice on the neonatal cases is alsoacknowledged.Mr. John Crumlish and Dr. Richard Hagan fortheir advice on serological issues and onlaboratory matters.Thanks also to:Dr. Joan Power, Dr. Nuala Moore and Dr. MichaelThomas of the MRTC, Ms. Mirenda O’DonovanHead of Corporate Affairs, Mr. Peter McDonnellTraining Officer, Ms. Niamh O’Sullivan, Ms. LucyO’Doherty and Ms. Janet Kelleher of LibraryServices and the staff of the IBTS IT Department.The staff of the Virology Laboratory, Ms. CarmelSheridan, Recipient Tracing Unit and Ms. PaulineCoakley Quality Assurance Manager NBC andDr. Jeff Connell and staff of the National VirusReference Laboratory, University College Dublinfor their assistance in suspected transfusiontransmitted donor investigations.The staff of the National Histocompatibility andImmunogenetics Reference Laboratory of theIBTS and Mr. Geoff Lucas and staff at theInternational Blood Group Reference Laboratory,National Blood Group Reference Laboratory,National Blood Service, Bristol, UK for their helpin the investigation of donors involved in TRALI.114

References:Ahmad, I. and Gibson, P.R., (2006) Management ofiron deficiency in patients admitted to hospital: timefor a rethink of treatment principles. Internal MedicineJournal, 36(6), 347–354.American Association of Blood Banks (2005) AmericanAssociation of Blood Banks Technical Manual, 15thEdition. Bethesda MD: AABB Press. (Editor- Mark E.Breecher).American Association of Blood Banks (2008) AmericanAssociation of Blood Banks Technical Manual, 16thEdition. Bethesda MD: AABB Press. (Editor- Roback, J,Combs, MR, Grossman, BJ. And Hillyer, C).Andrzejewski, C and Popovsky, M. (2005) Transfusionassociatedadverse pulmonary sequelae:widening ourperspective. Transfusion, 47 (7), 1048-1050Bosse, G., Breur, J. and Spies, C. (2006) The resistanceto changing guidelines- what are the challenges andhow to meet them. Clinical Anaesthesiology, 20(3),379-395.Bowman, J. (1977) The prevention of RhImmunization, Canadian Family Physician, 23, 60-68.British Committee for Standards in Haematology,British Transfusion Taskforce (1999) The administrationof blood and blood components and themanagement of transfused patients, TransfusionMedicine, 9 (3), 227-238British Committee for Standards in Haematology(2003) Guidelines for the use of platelets, BritishJournal of Haematology, 122(1), 10-23British Committee for Standards in Haematology,Blood Transfusion Task Force (2004) Guidelines for theuse of FFP, cryoprecipitate and cryosupernatant.British Journal of Haematology, 126 (1), 11-28.British Committee for Standards in Haematology(2006a) Guidelines for the use of prophylactic anti –Dimmunoglobulin. Available at:http://www.bcshguidelines.com/pdf/Anti-D_070606.pdf. [Accessed on: 26/07/2010].British Committee for Standards in Haematology(2006b) Guidelines for blood grouping and antibodytesting in pregnancy. Available at:http://www.bcshguidelines.com/pdf/Anti-D_070606.pdf. [Accessed on: 26/07/2010].British Committee for Standards in Haematology(2009) Guideline on the Administration of BloodComponents. Available at:http://www.bcshguidelines.com/4_HAEMATOLOGY_GUIDELINES.html?dtype=Transfusion&dpage=0&sspage=0&ipage=0#gl [Accessed on: 26/07/2010].British Committee for Standards in Haematology(2010) Guidelines on the use of irradiated bloodcomponents. Available at:http://www.bcshguidelines.com/publishedHO.asp?tf=Blood%20Transfusion&status=. [Accessedon:26/07/2010].Chaffe, B., Jones, J., Milkins, C., Taylor, C., Asher, D.,Glencross, H., Murphy, M. and Cohern H. (2009) UKTransfusion Laboratory Collaborative, recommendedminimum standards for hospital transfusionlaboratories. Transfusion Medicine, 19(4), 156-158.Commission on Patient Safety and Quality Assurance(2008) Building a culture of patient safety, report ofthe Commission on Patient Safety and QualityAssurance, Available at:http://www.cpsqa.ie/publications/ [Accessed on03.02.2009].Davenport, R. (2005) Blood components should belabelled for content, Transfusion. 45(1), 3-4.Davenport, R. (2005) Blood components should belabelled for content, Transfusion. 45(1), 3-4.British Committee for Standards in Haematology(2004). Transfusion guidelines for neonates and olderchildren. British Journal of Haematology 124,433-453Egan, L. Coneally, E., Lawlor, E. and Gough, D. (2010)P-1032 Red cell usage for iron deficiency anaemia in auniversity teaching hospital. Vox Sanguinis, 99 (Supp1), 1253.115

Gauvin, F., Lacroix, J., Robillard, J., Lapointe, H. andHume, H. 2006. Acute transfusion reactions in thepaediatric intensive care unit. Transfusion. 46 (11), 1899-1908.Gilcher, R.(2002) Apheresis: principles and technology;in Simon, T. Dzik, W. Snyder, E. Stowell, C. Strauss, R.(eds): Rossi’s Principles of Transfusion Medicine, 3rdEdition, Philadelphia, PA, Lippincott Williams & Wilkins,648-658.Goodnough,L. T., Manaitis, A., Earnshaw, P. and theNATA Consensus Development Working Group (2010)Management of preoperative anaemia in patientsundergoing elective surgery, ISBT Science Series. 5(1),120-124Heddle, N.M. and Kelton J. G. (2001) FebrileNonhaemolytic Transfusion Reactions In: Popovsky, MA.Transfusion Reactions 2nd Edition, Bethesda, MD:AABB Press, 45-83.Health Information and Quality Authority (2009)Recommendations for a Unique Health Identifier forindividuals in Ireland. Available at:http://www.hiqa.ie/media/pdfs/Unique_Health_Identifier_Report.pdf[Accessed on 08/10/10]Health Information and Quality Authority (2010) DraftNational Standards for Safer Better Healthcare [Online}Available at:http://www.hiqa.ie/safer_better_healthcare.asp[Accessed on 08/10/10]Kleinman, S., Caulfield, T., Chan, P., Davenport, R.,McFarland, J., McPhedran, S., Meade, M., Morrison, D.,Pinsent, T., Robillard, P., Slinger, P. (2004) Toward anunderstanding of transfusion –related lung injury:statement of a consensus panel. Transfusion. 44(12),1774-1789Kopko PM and Popovsky MA. (2007) Transfusion-Related Acute Lung Injury In Popovsky MA. TransfusionReactions, Third Edition, Bethesda, MD: AABB Press.207-228Kulnigg S, Stoinov S, Simanenkov V, Dudar LV, KarnafelW, Garcia LC, Sambuelli AM, D'Haens G, and Gasche C.(2008) A Novel Intravenous Iron Formulation forTreatment of Anemia in Inflammatory Bowel Disease:The Ferric Carboxymaltose (FERINJECT) RandomizedControlled Trial, American Journal of Gastroenterology;103(5), 1182-1192.Leape, L., Bates, D., and Cullen, D., Cooper J,Demonaco HJ, Gallivan T, Hallisey R, Ives J, Laird N,Laffel G, et al (1995) Systems analysis of adverse drugevents. Journal of the American Medical Association.274(1), 35-43.Lee, D. and Rawlinson, V. (1995) Multi centre trial of lowdose Anti-D immunoglobulin, Transfusion Medicine,5(1), 15-19.McClelland, DBL (Ed), (2001) Adverse effects oftransfusion In Handbook of Transfusion Medicine, ThirdEdition, United Kingdom: The StationaryOffice, 91-103.Hogman, CF. Meryman, HT. (2006) Red blood cellsintended for transfusion: quality criteria revisited.Transfusion, 46(1), 137-142.International Society of Blood Transfusion (2006)International Congress of the International Society ofBlood Transfusion (ISBT) Working Party XXIIX,Capetown, South Africa.McClelland, DBL (Ed), (2007) Handbook of transfusionmedicine, United Kingdom Blood services, 4th Edition.Available at:http://www.transfusionguidelines.org.uk/index.aspx?Publication=HTM [Accessed on: 26/07/2010].McDonald, CP. (2006) Bacterial risk reduction byimproved donor arm disinfection, diversion andbacterial screening. Transfusion Medicine. 16 (6), 381-396.116

Medical Council (2009) Guide to Professional Conductand Ethics for Registered Medical Practitioners 7thEdition. Available at:http://www.medicalcouncil.ie/Professional-Standards/Professional-Conduct-Ethics/.[Accessed on:26/07/2010].Mollison, PL. Engelfriet, CP. and Contreras, M. (1997)Blood Transfusion in Clinical Medicine, 10th Edition.Oxford, United Kingdom: Blackwell Science.Mollison, P., Engelfriet, C. and Conreras, M. (2005)Haemolytic Disease of the fetus and the newborn inMollisons Blood Transfusion in Clinical Medicines. (Klein, H and Anstee, D –Editors)Oxford: Blackwell PublishingLtd. 496-545.Murphy, M and Little, T. (2008) A survey of hospitalblood transfusion laboratory information technologysystems and their functionality. Transfusion Medicine.18(3), 204-206.National Haemovigilance Office (2007) NationalHaemovigilance Office Annual Report. Dublin: NationalHaemovigilance Office.National Institute for Clinical Excellence (NICE ) (2002)Pregnancy -routine anti-D prophylaxis for pregnantwomen. Technical appraisal No 41. London: NICE.National Institute of Clinical Excellence (2008) Routineante natal prophylaxis for women who are Rh Dnegative (review of NICE technology appraisalguidance 41. London: NICE.National Patient Safety Agency (2006) Safer practicenotice 14, right patient, right blood. Availableat:http://www.nrls.npsa.nhs.uk/resources/?entryid45=59805. [Accessed on: 05/11/2009]New, HV. (2006) Paediatric transfusion . Vox Sanguinis.90 (1), 1-9Napoli, A., Jagoda, A. (2007) Clinical policies, theirhistory, future, medical, legalimplications and growing importance to physicians,Administration of EmergencyMedicine, 33(4), 425-32.National Blood Users Group (2004) Guidelines for theadministration of blood and blood components.Dublin: National Blood Users Group.National Blood Users Group (2007) Transfusion ofBlood Components to Infants Under Four Months:Review and Guidelines. Irish Medical Journal 100 (6)(Supplement),1-22.National Haemovigilance Office (2004) NationalHaemovigilance Office Annual Report. Dublin: NationalHaemovigilance Office.National Haemovigilance Office (2006) NationalHaemovigilance Office Annual Report. Dublin: NationalHaemovigilance Office.Ohaska,A.(2009) Transfusion errors and theirprevention. ISBT Science Series 4(n2),390-394.O’Riordan,J.,Fitzgerald,J.,Smith,O.P.,Bonnar,J.,Gorman,W.A.(2007). Transfusion of blood components to infantsunder four months. Irish Medical Journal. 100 (6),Supplement1-23.Pirie, ES. and Gray, MA. (2007) Exploring the assessors’and nurses’ experience of formal assessment of clinicalcompetency in the administration of blood componentsNurse Education in Practice, 7(4), 215-27.Popovsky, MA. (2001) Circulatory Overload in Popovsky,MA. Transfusion Reactions Second Edition. Bethesda,MD: AABB Press, 255-259.Reason, J. (2000) Human error: models andmanagement, British Medical Journal,320(7237), 768-770.117

Scott, SD., Hirschinger, L. E,, Cox, K. R. McCoig, M.,Brandt, J. and Hall LW. (2009) The natural history ofrecovery for the healthcare provider “second victim”after adverse patient events. Quality and Safety inHealth Care, 18(5), 325-330.Valeri, C.R. Dennis R.C. Ragno, G. Macgregor, H.Menzoian, J.O. & Khuri, S.F. (2006) Limitations of thehaemacrit level to assess the need for red blood celltransfusion in hypovolemic anemic patients.Transfusion, 46(3), 365-371Senge, P. (1990) The fifth discipline, the art and practiceof learning organisations. New York: Doubleday.Serious Hazards of Transfusion Scheme. (1999) SeriousHazards of Transfusion (SHOT) Annual Report 1998-1999. Manchester: SHOT Office, (National Coordinators– Dr. Elizabeth M Love, Ms. Kate Soldan).Serious Hazards of Transfusion Scheme. (2009) SeriousHazards of Transfusion (SHOT) Annual Report 2008.Manchester: SHOT Office,Vincent, C., Taylor –Adams, S.,Chapman, E. J., Hewitt,D. Prior, S., Strange, P. and Tizzard, A. (2008) How toinvestigate and analyse clinical incidents; Clinical RiskUnit and Association of Litigation and RiskManagement protocol, British Medical Journal,320(7237), 777-781.Wu, A.W. (2000). ‘Medical Error: the Second Victim.British Medical Journal, 320(7237), 726-727.Soni, N. (2008) Intraoperative transfusion, triggers andprecision – an oxymoronTransfusion Medicine 18(4), 209-210Stainsby,D., Jones,H., and Casbard,A. et al.(2008).Adverse outcomes of blood transfusion in children:analysis of UK reports to the serious hazards oftransfusion scheme 1996-2005. British Journal ofHaematology 141(1), 73-80.The AGREE Collaboration. Writing Group: Cluzeau FA,Burgers JS, Brouwers M, Grol R, Mäkelä M, LittlejohnsP, Grimshaw J, Hunt C (2003) AGREE Electronic Libraryfor guideline developers. Available at:http://www.agreecollaboration.org/. [Accessed on:11/08/2009].The Office of Health Management (2004) Directory ofCompetency Development Options. Available at:http://www.hseland.ie/download/pdf/competency_directory.pdf. [Accessed on; 08/11/2009]Tovey, L., Townley, A., Stephenson, B. and Taverner, J.(1983) The Yorkshire antenatal Anti-D immunoglobulintrial in primegravidae. Lancet. 2(8344), 244-246.118

Appendix 1:Incidence of IBCT/SAE and SAR per Unit distributed from IBTS1. 2008Category Red Cell Platelets Plasma Granulocytes Cryoprecipitate TotalConcentrate 24,624 24,330 285 2,717 components(RCC)issued144,383 from IBTS196,339IBCT/SAE (111) 1 per 1,769Febrile Non Haemolytic 1 per 4125 1 per 1 per 285 1 per 5,167Transfusion Reaction (38) 1 12,312Immunological Haemolysis 1 per 72,192 1 per 98,170due to other alloantibody(Acute

Appendix 1:2. 2009Category RCC/ Platelets Plasma Granulocytes Cryoprecipitate TotalWhole Blood 26,328 23,876 328 1,196 components146,584 distributed/issuedfrom IBTS198,355 8IBCT/SAE (157) 1 per 1,263FNHTR (37) 1 1 per 4581 1 per 6582 1 per 5,361 7Immunological Haemolysis 1 per 73,292 1 per 99,178due to other alloantibody(Acute

Appendix 2:Classification of mandatory reports 2008 (n=55)SAE - STEP IN WORK PROCESSSPECIFICATIONTotal Product Equipment Human Otherdefect failure error specifyWhole Blood Collection 0 0 0 0 0Apheresis Collection 0 0 0 0 0Testing of Donations 1 0 0 1 0Processing 2 0 0 2 0Storage 10 0 2 8 0Distribution 2 0 0 1 1Materials 0 0 0 0 0Other - Transfusion of IncorrectlyLabelled Component 13 0 0 12 1Other - Non-Irradiated/CMV NegComponents Transfused 3 0 0 3 0Other - Incorrect ABO groupTransfused (No Reaction) 2 0 0 2 0Other - Incorrect ComponentTransfused (No Reaction) 4 0 0 3 1Other - Transfusion of Other AntigenIncompatible Component (No Reaction) 6 0 0 3 3Other - Incorrect Rh groupTransfused (No Reaction) 4 0 0 4 0Other - Transfusion of ExpiredComponent 4 0 0 4 0Other - Error occurring in BEnot covered above 1 0 0 1 0Other (Specify) 3 1 0 1 1121

Appendix 3:Nature of adverse event categorised as "Other" 2008, (n=17)Category n Mandatory Non MandatorySAESAEInappropriate component transfused -Patient inSickle Cell Crises transfused with ABO and RhDcompatible red cells, but units were not phenotyped 1 1 0Incorrect giving set used red cells and platelets. 3 0 3Red cells transfused with incorrect unit number – defaultnumber on printer not adjusted to correct unit number. 1 1 0Incorrect transfusion time 7 0 7Other- transfusion of red cells where there were clotscontained in pack.Unfortunately, the pack was discarded, and follow-upinvestigations on the pack could not be completed. 1 1 0Other –Transfusion of red cells where pack was perforated. 2 0 2Other - Patient transfused uncross- matched unit of red cells. 1 1 0Other - Patient commenced on antibiotics following error inthe BE and hospital blood transfusion laboratory. 1 1 0122

Appendix 4:Description of root cause codesHUMAN FAILURECONTRIBUTINGFACTORSDESCRIPTIONVerificationErrors which occur following incomplete assessment of a situation includingrelated conditions of the patient/donor and materials to be used beforebeginning a task. Examples would be failure to obtain positive patient ID at thebedside, failure to verify most recent test results prior to prescribing and failureto verify current results against historical results where applicable.KnowledgeAn error occurs when the individual is unable to apply their existing knowledgeto a novel situation. Examples would be a trained medical scientist who isunable to solve a complex antibody issue, or a trained nurse who fails to takeinto account the patient’s blood group prior to commencing a transfusion.Co-ordination /CommunicationAn error occurs due to a lack of communication or co-ordination within a teamfor example where an intention to cancel a prescription is not communicatedresulting in the patient receiving an inappropriate transfusion.MonitoringErrors occur where there is a failure to monitor a process or patient status.Examples include failure to monitor rate of transfusion leading to patient beingtransfused too quickly/slowly or a trained medical scientist operating anautomated instrument and not realising that the pipette dispensing the reagentis clogged.SlipErrors occur where there is a failure in the performance of highly developedskills for example computer entry error or simple mind slip leading to failure tocomplete a task.TripFailures in whole body movement, for example dropping a blood bag whichsplits and is wasted.Patient RelatedErrors which occur directly as a result of actions or characteristics of patients,and which are not within the control of the health care team. Examples includewhere a patient gives wrong information about their patient details or wherepatients remove their own ID band.UnclassifiableErrors which arise and cannot be classified in any of the current categories.123

SYSTEM FAILURECONTRIBUTINGFACTORSDESCRIPTIONDesignErrors which arise due to inadequate design of equipment, software ormaterials e.g. design of workspace, software packages, or label design.MaterialsErrors which arise due to deficits in materials e.g. defects in label adhesive, orink smears on pre-printed labels or forms.ConstructionErrors which occur following poor construction e.g. incorrect set-up of bloodpumps/laboratory equipment or installation of equipment in an inaccessiblearea.Management PrioritiesErrors which occur as result of organisational management prioritisation of otherissues over safety e.g. decisions on staffing levels, limited or absentphlebotomy services, no provision for medical record numbers out of hours(Lundy et al, 2007)Policies andProceduresErrors which occur due to unclear/outdated or absent Standard OperatingProcedures (SOP)Policies/procedures should be current, understandable well presented andaccessible to all staff.CultureErrors which arise from a collective approach to safety and risk. Groups mayestablish their own modes of function as opposed to following prescribedmethods e.g. not paging a manager/doctor out of hours to review a result/decision, as it is not usual practice.(Available at http://www.mers-tm.org/, Accessed on 21/12/2009)124

NOTES:125

126NOTES: