SEARCH

SEARCH BY CITATION

Keywords:

  • Anti-D immunoglobulin;
  • errors;
  • haemolytic disease of the fetus and newborn;
  • pregnancy;
  • prophylaxis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Objective

To highlight the errors associated with the use of anti-D immunoglobulin in RhD antigen-negative women, and their resultant clinical impact during and after pregnancy, and to suggest strategies to reduce these errors.

Design

Retrospective review of cumulative reporting to the UK confidential haemovigilance scheme, Serious Hazards of Transfusion (SHOT), between 1996 and 2011.

Setting

Obstetric departments in the UK.

Population

Mothers who require anti-D immunoglobulin to prevent RhD sensitisation during pregnancy or after birth.

Methods

Hospital transfusion teams reported adverse events to the SHOT database.

Main outcome measures

Reported number of events and their causes, and morbidity and mortality associated with errors.

Results

In 15 years of reporting, SHOT haemovigilance has shown a total of 1211 errors related to the administration of anti-D immunoglobulin, particularly regarding omission or late administration (157/249 or 63% reported in 2011). Anti-D immunoglobulin errors comprised 13.7% (249/1815) of all SHOT reports in 2011. Failure to recognise women who already have RhD sensitisation occurred in 19 cases, and was followed by suboptimal monitoring of the pregnancy. Nine of the infants suffered haemolytic disease of the fetus and newborn (HDFN): one resulted in neonatal death and three required red cell transfusion.

Conclusions

Babies as well as their mothers remain at risk from avoidable errors. More active attention at national and local levels to further education and training, particularly for midwives, is an absolute necessity. We recommend the use of a SHOT-devised anti-D administration flowchart, adapted locally into a checklist, to help reduce errors.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Extensive literature demonstrates that patients are at risk of harm from errors made by those caring for them. Many errors can be prevented by putting in place appropriate checks. Haemolytic disease of the fetus and newborn (HDFN) is a preventable cause of death and serious morbidity. Anti-D immunoglobulin (anti-D Ig) can prevent RhD antigen-negative women developing immune anti-D, which is responsible for more than 90% of the cases of HDFN. This has been generally very successful, but here we report potentially avoidable errors that continue to put babies, as well as their mothers, at risk.

The introduction of postnatal anti-D Ig in the UK in 1969 has led to a dramatic reduction in the number of fetal and neonatal deaths related to RhD alloimmunisation, and to a corresponding fall in the rate of anti-D alloimmunisation. Routine antenatal anti-D prophylaxis (RAADP) has been recommended at 28 and 34 weeks of gestation in the UK for all RhD-negative women, in order to prevent sensitisation by small ‘silent’ undetected bleeds until delivery.[1, 2] This has further reduced sensitisation. With effective RAADP readily available, it is the right of every RhD-negative mother and her baby to receive this treatment to prevent the potentially devastating consequences of maternal sensitisation to the RhD antigen.

Compliance with current national guidance for the administration of anti-D Ig is key to the avoidance of sensitisation.[2, 3] Effective prevention of RhD sensitisation involves several steps, including the decision to prescribe anti-D Ig, laboratory testing, and issuing and administering anti-D Ig to the correct patient in a timely manner, and these steps involve several healthcare professionals in different disciplines. Errors can occur at any stage of the process, and a failure to recognise potentially sensitising events during pregnancy and a failure to manage them appropriately when they do occur are the main causes. The current incidence of RhD sensitisation is not known, and is not captured by existing national reporting systems. Although the number of perinatal deaths is recorded, there are no published data about deaths related to HDFN.

The Serious Hazards of Transfusion (SHOT) UK-wide confidential reporting system, established in 1996, collects data on all adverse events and reactions related to the transfusion of labile blood components. No patient-identifiable information is submitted, so ethical approval is not required. Initially voluntary, SHOT reporting is now required for quality and safety monitoring by the Departments of Health in England, Scotland and Wales, and for clinical laboratory accreditation within the National Health Service (NHS) of the UK. Annual SHOT reports included adverse events related to anti-D Ig from the outset. With the increasing recognition of harm, such events were actively sought from 2006 onwards. We here review SHOT data from 1996 to 2011 relating to anti-D Ig administration errors. Allergic and other reactions are reported to the Medicines and Healthcare products Regulatory Agency (MHRA), and are rare, with only 93 adverse reactions (none fatal, seven anaphylactic) reported between 1991 and August 2012.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Hospital staff, the majority being transfusion practitioners or biomedical scientists (BMSs), reported adverse events to SHOT using an online computer database. In the first decade (1996–2005) reporters telephoned the SHOT office and were mailed a paper questionnaire to complete. In January 2010 online reporting became available (Dendrite Clinical Systems™). All anti-D reports were analysed for reliability by SHOT staff, and the reporter was contacted if necessary for clarification. As this was initially a voluntary reporting scheme, and because hospitals varied in their case ascertainment rates, anti-D incidents reporting was incomplete, but has increased with time as error reporting has become professionally mandated.

The anti-D reporting categories are listed below

  1. Omission or late administration of anti-D Ig (>72 hours following an event).
  2. Inappropriate administration of anti-D Ig to an RhD-positive patient, a patient who already had immune anti-D, a mother of an RhD-negative infant, or administration to the wrong patient.
  3. An incorrect dose of anti-D Ig, according to local policy.
  4. Anti-D Ig handling and storage errors.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

There were 1211 errors related to anti-D Ig reported between 1996 and 2011, with a steady increase in reports each year. In 2011, with 98.4% of all UK hospitals/trusts submitting reports, anti-D Ig errors accounted for 249 (13.7%) of the total of 1815 SHOT cases. Of these, 63% (157/249) were late or missed doses in 2011. These also accounted for nearly 50% (609/1211) of all cases in the reporting period. The cumulative numbers of anti-D Ig errors are shown in Table 1. The majority (842/1211; 69.5%) of errors were attributable to nurses or midwives. Overall, clinical errors were responsible for 72.5% (879/1211) of all reports, and laboratory errors were responsible for the remainder (332/1211; 27.5%).

Table 1. Incident types and initial mistakes 1998–2011
Type of eventNumber of reports
CasesInitial mistake made by:
Nurse/midwifeLaboratoryDoctor
Omission or late administration of anti-D Ig6095266122
Anti-D Ig given to RhD-positive mother2801531189
Anti-D Ig given to mother with immune anti-D10864440
Anti-D Ig given to mother of RhD-negative infant6114470
Anti-D Ig given to wrong patient494702
Wrong dose of anti-D Ig given5416362
Anti-D Ig handling and storage errors5022262
Total

1211

(100%)

842

(69.5%)

332

(27.5%)

37

(3.0%)

Omission or late administration of anti-D Ig

In 548/609 cases (90%), the primary error or omission occurred in the clinical area. Lack of communication and poor documentation were common features. A key finding was a failure of the maternity discharge check. Inadequately labelled blood samples, with a delay in receiving repeat samples, also caused delays in the administration of anti-D Ig.

Anti-D Ig given to an RhD-positive mother

This occurred in 280/1211 cases (23%), with the primary error occurring in the clinical area in 162/280 (58%) cases, and in the laboratory in 118/280 (42%) cases.

Anti-D Ig administered to mothers with immune anti-D

These comprised 9% of all anti-D incidents. Most cases (64/108; 59%) occurred in the clinical area. Serious errors included misinterpretation of laboratory results, with anti-D antibodies attributed to passive prophylaxis rather than recognition that this was maternal immune anti-D. The pregnancies were then not appropriately monitored, resulting in unanticipated HDFN. One mother was found to have developed a strong anti-C+D response, and the baby suffered from HDFN, dying 3 days after an exchange transfusion. Overall, there were 19 cases where anti-D Ig was administered to mothers with immune anti-D, resulting in nine cases of HDFN, three of which occurred between 2005 and 2010; six cases occurred in 2011, one of which required an intrauterine transfusion, and two others required top-up transfusions.

Other mistakes included anti-D Ig being given to the mother of an RhD-negative infant in 61 cases (5%), with 47/61 (77%) of these mistakes originating in the laboratory. Anti-D Ig was given to the wrong patient in 49 (4%) cases as a result of a failure in basic checking procedures prior to administration. There were 54 cases where the wrong dose of anti-D Ig was given (4.4% of total errors), with 66.5% of these errors originating in the laboratory, where a BMS issued a dose of immunoglobulin that was inappropriate to the clinical situation.

Anti-D Ig handling and storage errors

There were 50 errors (4.0%) in handling, including the issue of expired anti-D Ig from ward stocks, and laboratory errors involving inadequate recording of batch numbers. In other cases completely different medicinal blood products (such as anti-tetanus Ig) were given instead of anti-D Ig, and vice versa.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Main findings and interpretation

Preventable drug errors are well-recognised, and we observed that over the 15 years of SHOT reporting on anti-D Ig, the same mistakes were being made repeatedly by clinical and laboratory staff. Of a total of 1211 errors, 879 (72.5%) involved clinical staff, and 50% of the total (609/1211) were an omission or late administration of anti-D Ig, which had the greatest potential for clinical harm as the woman may become sensitised and develop anti-D, putting future pregnancies at risk. Errors were caused by the failure to follow basic protocols, failure to take into account laboratory computer records, poor communication and poor decision-making, compounded by poor understanding, such as the misinterpretation of the significance of anti-D (assuming it was a consequence of prophylaxis rather than a new immune event). As a result, there were 19 cases where an anti-D antibody was wrongly attributed to passive prophylaxis: nine babies suffered HDFN (six in 2011), with one neonatal death, and three babies requiring red cell transfusion. The magnitude of harm as a result of anti-D Ig errors is not known, mainly because the adverse outcomes for both mother and baby following these errors may be delayed. A SHOT-requested search in June 2012 on perinatal deaths by the Office of National Statistics (ONS) found 12 stillbirths and 22 neonatal deaths (total 34) that included ‘rhesus isoimmunisation’ on the death certificate in England and Wales between 2001 and 2010 (ONS, pers. comm.). This suggests preventable deaths are still occurring, and that complacency about the success and effectiveness of anti-D Ig treatment is misplaced.

Strength and weaknesses

SHOT data have weaknesses: reporting is professionally mandated, but is not a legal requirement; there are few denominator data, and no means of verifying the accuracy. These reports may be the tip of the iceberg, as SHOT benchmarking demonstrates considerable variation in the level of hospital reporting of adverse events.[4] In 2010, 92 (48%) of NHS trusts reported anti-D events. SHOT collects data on the number of doses of anti-D Ig administered by hospitals who report anti-D events, and for 2010 there were 256 reports for 141 803 doses of anti-D Ig, giving an estimated event rate of 1.8 errors per 1000 doses. Additional data provided by anti-D Ig suppliers support this figure. Anti-D Ig errors are not reported in many other haemovigilance schemes, apart from in Ireland, where the errors recorded show a similar pattern.[5] Despite the limitations, SHOT data provide clear evidence of anti-D Ig errors, with demonstrable and potential adverse clinical impact, and should therefore be used to underpin strategies to reduce avoidable anti-D Ig errors.

Implication for clinical practice

The Royal College of Obstetricians and Gynaecologists and the British Committee for Standards in Haematology have published guidance on the use of anti-D Ig,[2, 3] and recommendations for safer practice have been published in annual SHOT reports.[6-8] The most recent health service circular Better Blood Transfusion mandates that the issue of anti-D Ig should follow the same stringent criteria with regard to patient identification and traceability as for any labile blood component.[9] Anti-D Ig is a prescription-only medicine, and therefore must only be administered after individual prescription, either by a medical officer or independent prescriber, under the auspices of a patient group directive, or under midwifery exemption rules.

Several initiatives have been developed in the UK to educate healthcare professionals about the use of anti-D Ig. These include e-learning modules added to the nationally adopted Learn Blood Transfusion e-learning package (www.learnbloodtransfusion.org.uk). The NHS Blood and Transplant Appropriate Use of Blood anti-D working group has developed posters and other educational material aimed at midwives and obstetricians;[10] however, the most recent (2011) SHOT data show no evidence of improvement in practice,[4] and further measures are required within obstetric departments to improve patient safety.

Maternal D-immunisation is now thought to be rare, which can lead to the erroneous assumption that all anti-D found in pregnancy is a result of previous treatment with anti-D Ig. This is dangerous, as misinterpretation of laboratory results can have potentially serious clinical implications: in half the cases here there was a failure to consult historical records for women with known previous anti-D. RhD typing should be performed by the routine methodology available in the transfusion laboratory, not by emergency techniques alone, which may not be as robust.

The issue of anti-D Ig from remote stocks held in the clinical area poses particular risks of inappropriate or incorrect administration. These errors led to 21% of all anti-D cases in 2008,[8] which suggests that there are insufficient checks in place to ensure safety and traceability.

Although some maternity units will require a local stock of anti-D Ig, its release should be subject to laboratory control, with a robust protocol.

Solutions

Two key steps are likely to lead to a reduction in anti-D Ig errors. First, at a national level, attention to further education and training is urgently required, particularly for midwives. The second key step is the implementation of a checklist. Several individuals across professional and managerial boundaries participate in the anti-D Ig administration pathway, and a lack of continuity is a recognised risk factor for error. The implementation of a checklist would be likely to lead to improved safety, as has been clearly demonstrated for surgery.[11] We recommend the use of a checklist developed from the SHOT-devised flowchart in Table 2, and modified for local use, to remind staff not to forget key steps. Accurate record-keeping of the use of anti-D Ig is hampered because it is a ‘batch’ product without individual identification numbers. Emerging technology, including radio-frequency ID chips attached to the box or vial, would permit secure, accurate traceability from the issuing department to the patient. Electronic patient records linked to laboratory reports including D-group would help, but individual obstetric departments should review and audit their procedures.

Table 2. Anti-D administration flowchart (reproduced with permission from SHOT, www.shotuk.org/resources)
  1. a

    FMH Fetomaternal haemorrhage.

Always confirm

 the woman's identity

 that the woman is RhD negative using the latest laboratory report

 that the woman does not have immune anti-D using the latest laboratory report

 that informed consent for administration of anti-D Ig is recorded in notes

Potentially sensitising events (PSEs) during pregnancy
Gestation LESS than 12 weeks
Vaginal bleeding associated with severe pain

Administer at least 250 iu anti-D Ig within 72 hours of event.

Confirm product/dose/expiry and patient ID before administration

ERPC/instrumentation of uterus
Medical or surgical termination of pregnancy
Ectopic/molar pregnancy
Gestation 12–20 weeks
For any potentially sensitising event

Administer at least 250 iu anti-D Ig within 72 hours of event.

Confirm product/dose/expiry and patient ID before administration

Gestation 20 weeks to term

For any potentially sensitising event

(Irrespective of whether RAADP has been given)

Request a Kleihauer test (FMHa test) and immediately administer

at least 500 iu anti-D Ig within 72 hours of event.

Confirm product/dose/expiry and patient ID before administration

Does the Kleihauer/FMH test indicate that further

anti-D Ig is required?

Administer more anti-D Ig following discussion with laboratory
 For continuous vaginal bleeding, at least 500 iu anti-D Ig should be administered at a minimum of 6-weekly intervals, irrespective of the presence of detectable anti-D, and a Kleihauer/FMH test should be requested every 2 weeks in case more anti-D is needed
Routine antenatal anti-D prophylaxis (RAADP)

For routine antenatal Anti-D prophylaxis

(Irrespective of whether anti-D Ig has already been given for PSE)

Take a blood sample to confirm group and check antibody screen.

Do not wait for results before administering anti-D Ig

Administer 1500 iu anti-D Ig at 28–30 weeks
OR

Administer at least 500 iu anti-D Ig at 28 weeks and

then administer at least 500 iu anti-D Ig at 34 weeks

Confirm product/dose/expiry and patient ID before administration
At delivery (or on diagnosis of intrauterine death >20 weeks)

 Is the baby's group confirmed as RhD positive?

OR

Are cord samples not available?

Request a Kleihauer test (FMHa test)

Administer at least 500 iu anti-D Ig within 72 hours of delivery

Confirm product/dose/expiry and patient ID before administration

Does the Kleihauer/FMH test indicate that further

anti-D Ig is required?

Administer more anti-D following discussion with laboratory

Unanswered questions and further studies

With this study, based on variable and voluntary reporting, SHOT has identified a clinically significant problem with anti-D Ig medication errors. The true magnitude of the problem is not yet known; however, a national audit of anti-D Ig by the National Comparative Audit of Blood Transfusion programme is planned for Spring 2013.[12] Revised guidelines on the use of anti-D Ig are currently in progress under the transfusion task force of the British Committee for Standards in Haematology, which will incorporate the anti-D Ig flowchart. In addition, from January 2013, SHOT have started collecting reports of women who present in pregnancy with a new immune anti-D antibody, and will learn whether some of these cases have resulted from late or missed administration of anti-D Ig in previous pregnancies.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

The SHOT analysis of reported anti-D Ig events has demonstrated clinically significant problems related to poor practices, which require active attention at national and local levels to further education and training, particularly for midwives, and remedial action by obstetric departments, including the use of a checklist to help reduce errors.

Disclosure of interests

All authors declare they had no support from any organisation for the submitted work, no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years, and no other relationships or activities that could appear to have influenced the submitted work.

Contribution to authorship

DP, and TD, collated and analysed the data. PBM, TD, and HC wrote the article. HC led the initiation and design of the data collection and contributed to the analysis over the full 15 years.

Funding

The SHOT haemovigilance scheme is funded by the UK Forum (the four blood services for the UK); no specific funding was received for this analysis.

Acknowledgements

The authors thank the reporters in hospitals and trust boards across the UK who have contributed data to SHOT.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References