Printer Friendly

A closer look at fatal transfusion reactions.

All U.S. health care facilities performing transfusions of blood or blood products are required to report any fatal transfusion reactions to the Food and Drug Administration. In a 1980 memorandum of understanding with the FDA, the Health Care Financing Administration agreed to investigate these fatalities in greater depth. The goal was to identify the causes of fatal reactions and develop guidelines for preventing their recurrence. This article presents the results of that research from its initiation in 1976 through 1983, the last year for which analyzed data are available.

First, an important preface: The number of deaths reported to the FDA as transfusion-related constitutes only a tiny fraction of the total number of units transfused. With 70 million units transfused over this seven-year period, the number of deaths deemed preventable is roughly one in every million units--an extremely small error rate.

Admittedly, this rate excludes any transfusion errors that did not result in death, or transfusion deaths attributed to other causes and left unreported. On balance, however, blood transfusion remains among the safest medical procedures. The HCFA follow-up reporting system aims to make it even safer.

These follow-up studies were a joint effort by both agencies. The FDA eviews all reports of transfusion fatalities and sets aside any reports from Medicare-funded facilities that appear to warrant further investigation. (From 1976 through 1983, FDA received approximately 275 fatality reports and earmarked some 20 a year for follow-up by HCFA.) Then HCFA conducts an on-site inspection to insure that the udnerlying cause of each incident is corrected. HCFA regional offices or appropriate state agencies provide the survey teams, which usually consist of a laboratorian, a nurse, and experts in other relevant areas such as medical records.

In order to remain in the Medicare program, facilities under investigation must show a plan to correct deficiencies. Surveyors use applicable hospital regulations and inspection documents, including FDA blood bank report forms. The survey findings and correction plans are sent to the facilities themselves, HCFA, and FDA. The Joint Commission on Accreditation of Hospitals and the American Osteopathic Association also receive survey information on hospitals in their accreditation programs.

The survey team reports its findings and any corrective actions taken. When facilities have failed to correct serious deficiencies. HCFA may provide them with consultation--for example, recommending that procedure manuals be improved and better qualified staff members hired.

Certain limitations apply when interpreting this information. The data in Figures I to III, based on reports submitted by hospitals and blood banks to the FDA, do not represent the actual number of transfusion fatalities. Each facility is responsible for determining a patient's cause of death. Therefore, a certain proportion of transfusion-related deaths may go unreported deliberately or inadvertently, due to legal concerns or misinterpretation of medical evidence. FDA and HCFA receive these reports without independent verification, and in some cases it is unclear whether the death resulted directly from transfusion or was merely incidental.

The major cause of transfusion-related death, according to these reports, was the administration of ABO-incompatible blood, accounting for 42 per cent of all fatal reactions reported (Figure I). The second leading cause was hepatitis (21 per cent); third was the presence of other antibodies that led to a hemolytic reaction (15 per cent).

In several instances, non-A, non-B hepatitis was the presumptive cause based on symptoms, test results, liver biopsy, or autopsy findings. Not all these factors were evaluated in every case, however. Since there is no definitive diagnostic test for non-A, non-B hepatitis, the condition was not always clearly established as the actual cause of death.

Currently, 5 to 18 per cent of Americans who receive five or more units transfused blood develop non-A, non-B hepatitis, according to a recent Congressional report by the Office of Technology Assessment. About 90 per cent of all post-transfusion cases of hepatitis are non-A, non-B, while some 10 per cent are due to hepatitis B.

Almost 57 per cent of all fatal transfusion reactions reported to the FDA were considered preventable, according to the study. The two leading types of preventable fatality, as shown in Figure II, are ABO incompatibility, accounting for 74 per cent, and antibodies undetected because of testing error, the cause of 21 per cent.

In this context, the term "preventable" is somewhat arbitrary. For example, is human error ever 100 per cent preventable? HCFA researchers use the word to mean that action based on good laboratory or medical practice--coupled with adherence to standard policies and procedures--could have kept ht edeath from occurring. Deaths due to non-A, non-B hepatitis, undetectable bacterial contamination, heart attacks, donor-related problems, or untraceable causes are deemed non-preventable.

The errors responsible for these preventable deaths are listed by hospital department in Figure III. The laboratory was the single department responsible for the largest proportion of these errors, including labeling mistakes; patient or specimen misidentification; incorrect grouping, crossmatching, or antibody identification; and clerical erros.

Patient misidentification, by the lab or another service, accounted for almost a third of all preventable deaths. Misidentified specimens or blood caused 8 per cent of such deaths, and technical errors in lab testing, 11 per cent. Another problem not covered in these reports is the failure to identify and treat a suspected transfusion reaction quickly.

Certain risk factors for fatal reactions become evident after reviewing the literature, HCFA regional studies, FDA reports, and follow-up data. These factors should be identified, reviewed, and corrected in any fatal transfusion incident. They should also serve as a cautionary checklist for any institution performing transfusions. Here is a rundown of the errors that can cost a life:

* Mislabeling of type and crossmatch requisition.

* Mislabeling or failure to label blood specimens.

* Improper patient identification when obtaining specimen or giving transfusion; misidentification of patients with similar names in the same ward, or of patients without ID bands who are unable to communicate (e.g., operating room, intensive care unit).

* Transcription of one patient's information onto the card and requisition of another.

* Failure to verify blood grouping or to compare information on the blood unit with medical records and patient ID.

* Taking the wrong blood units out of the operating room refrigerator.

* Blood unit mixup when two or more patients require blood simultaneously.

* Incompatible blood sent from the laboratory due to technical errors in the crossmatch.

* Possible ABO mistyping or failure to identify an antibody.

* Incorrect blood release form presented to blood bank, or release of wrong unit by blood bank.

* Administration of contaminated blood, or failure to recognize bacterial contamination.

* Administration of blood positive for hepatitis B, or suspected positive for non-A, non-B hepatitis.

* Improper preparation, storage, or transportation of blood and blood components resulting in contamination or deterioration.

* Failure by the medical or nursing staff to identify a transfusion reaction quickly.

* Failure to follow established procedures and policies.

* Inadequately trained personnel; lack of in-service training for transfusion personnel, including house, nursing, laboratory, and clerical staff.

* Inadequate staffing or supervision in the laboratory or on floors.

* Lack of follow-up by the facility after a transfusion death to prevent recurrence of the underlying problem.

What were the results from the HCFA follow-up investigations? In about 88 per cent of the cases, facilities corrected problems promptly, and the survey team merely verified that effective action had been taken. Figures IV and V summarize the problems identified by the follow-up teams and the actions taken, based on 59 cases studied between 1981 and 1983.

Floor error by nurses and other staff members caused half of the fatalities that were the subject of follow-up investigation, as Figure IV shows. These errors included inadequate staffing, clerical errors, and misidentification of patients and specimens. In a fifth of the cases, the fatal reaction turned out to be unrelated to transfusion, caused by factors ranging from an allergic reaction to the use of opiates. Laboratory error accounted for 18 per cent of the fatalities, and house staff error for 4 per cent.

Several hospitals had inadequate staffing on patient floors, undertrained and understaffed laboratory services, or house staff inadequately trained in transfusion reaction procedurs. Corrective actions, shown in Figure V, included provision of in-service training, begun by 24 per cent of the facilities investigated; changes in record keeping or operating procedures, 22 per cent; or a combination of these factors, 12 per cent. Employee dismissal or resignation accounted for 22 per cent of the actions. In 5 per cent of the cases, no correction was attempted since death resulted either from random human error or from a factor other than transfusion.

Using the results of this study, HCFA and its regional offices are working with the Food and Drug Administration and several private organizations to develop procedures and guidelines for investigating and preventing fatal reactions. The guidelines for investigating and preventing fata reactions. The guidelines are scheduled for release this year.

While blood transfusion remains an extremely low-risk procedure, these statistics remind us that one preventable fatality is too many. It is encouraging to note that, according to this survey, most facilities do everything possible to guard against recurrence of fatal reactions.
COPYRIGHT 1985 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1985 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Edinger, Stanley E.
Publication:Medical Laboratory Observer
Date:Apr 1, 1985
Previous Article:'Sorry, John, we have to let you go'; encouragement and placement assistance help laid-off employees and make remaining employees feel better about...
Next Article:Are you acting or just pretending? Interim management appointments can benefit the laboratory when they are used to groom potential promotables.

Related Articles
What every laboratorian should know about AIDS.
Making home transfusions work.
Achieving better blood bank QA with a transfusion form.
Hepatitis C test shows promise, pitfalls.
A phased approach to transfusion reaction investigation.
Banking on blood conversion: new technology may change the character of the U.S. blood supply.
Converting Stored Blood.
Deadly blood transfusion problem is preventable. (Blood Bank).
Transfusion safety: what has been done; what is still needed?
Who should initiate a transfusion-reaction investigation?

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters