| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology,
Department of Transfusion Medicine and Blood BankUniversity of Cologne, Germany
Address correspondence and reprint requests to Jens Krombach, MD, Department of Anesthesiology, University of Cologne, 50924 Cologne, Germany. Address e-mail to krombach{at}netcologne.de
 |
Abstract |
|---|
 Top Abstract Case 1DiscussionReferences |
|---|
IMPLICATIONS: Human error leading to the transfusion of blood to an unintended recipient is a major source of transfusion-related fatalities. We report five cases that highlight some specific areas in which transfusion error is likely to occur.
|
Case 1 |
|---|
|
TopAbstractCase 1DiscussionReferences |
|---|
Case 2
A multiple trauma patient, accompanied by the physician of the ambulance vehicle, was admitted to the emergency room (ER). The ambulance physician handed an unlabeled blood sample "of this patient" to the attending anesthesiologist of the ER. The anesthesiologist labeled the blood tube with the patient’s name and filed a request for typing and crossmatching to the blood bank. Twenty minutes later, four units of packed red cells were brought to the ER. The anesthesiologist performed a bedside ABO compatibility test with blood freshly drawn from the patient’s vein. The test result was "O Rh positive," whereas the blood bank had determined "B Rh positive" and had delivered B Rh positive packed red cells accordingly. Retyping and crossmatching with another sample of the patient’s blood in the blood bank confirmed the result of the bedside pretransfusion compatibility test "O Rh positive." As to subsequent investigation, the blood sample provided by the ambulance physician originated from a second victim of the same accident whom he had already taken to a different hospital. However, the attending physician of the other hospital had refused to accept the blood sample, as it was unlabeled.
Case 3
During dramatic emergency thoracotomy with extracorporeal circulation for massive intrathoracic hemorrhage, the anesthesiologist ordered ten units of packed red cells, with highest priority and without crossmatch, in addition to eight units that were already available in the operating room. The patient’s blood group was known to be "O Rh positive." Without checking the labels of the blood bags, the anesthesiologist rapidly transfused the first eight units, trusting that they were all "O Rh positive." Eventually, it was discovered that two of the eight units were in fact "A Rh positive," and that one of them had already been transfused. Surgery was completed successfully, and the patient remained asymptomatic with respect to the transfusion error. Massive hemorrhage and hemodilution (i.e., loss and dilution of the recipient’s A and B antibodies against the donor red cells) may have prevented a hemolytic transfusion reaction.
Case 4
By the end of the regular day shift of an intensive care unit (ICU), the anesthesiologist on call took over from his colleague about to leave. As part of the remaining agenda, a first unit of packed red cells was to be transfused to a postoperative patient. The identification procedure and pretransfusion ABO compatibility test with the patient’s blood was already performed and signed by the colleague from the day shift, and the records were ready for inspection in the ICU nurses’ room. There, the anesthesiologist on call repeated the identification procedure by comparing his colleague’s records with the information on the form attached to the blood unit to be transfused. He then entered the patient’s room with the unit of red cells prepared for transfusion. The room was shared by two patients with one’s last named being the same as the other’s first name, such as Eric Walter versus Walter Paul. The blood was transfused to the wrong patient, who had a different blood group and who developed hemoglobinuria approximately half an hour later. Fortunately, there were no further complications and the outcome was favorable.
Case 5
Two patients with identical first and surnames but different dates of birth underwent similar noncardiac vascular surgery on close dates. Preoperatively, patient A had undergone autologous blood donation. Patient B required postoperative blood transfusion. With no difference in name, the autologous blood of patient A was filed to patient B by a nurse on the ward. The attending physician checked the accompanying identification material, however, without paying attention to the dates of birth recorded on the patient’s chart and the form attached to the unit to be transfused. He did not deem these details very important because it was "just autologous blood." For the same reason, he also refrained from pretransfusion ABO compatibility testing. After transfusion of the first 100 mL of red blood cells the patient developed acute hemolysis and shock. He died on the same day. The case went to court.
|
Discussion |
|---|
|
TopAbstractCase 1DiscussionReferences |
|---|
In the United States, the frequency of avoidable transfusion fatalities attributable to misidentification of the pretransfusion blood sample, the blood unit, or the recipient has been reported to range from 1 per 600,000 to 1 per 800,000 transfusions (2,6). The estimated incidence of nonfatal transfusion errors in the United States ranges from 1 in 12,000 to 1 in 19,000 transfusions (2,6).
The United Kingdom’s Serious Hazards of Transfusion (SHOT) group, which tracks adverse events on a voluntary basis and currently covers more than 90% of all red cell usage in the United Kingdom, recently reported an error incidence of 1 in 16,000 (335 errors per 5.5 million units of red blood cells transfused between 1996 and 1999) (10). Transfusion errors accounted for 54% of all transfusion-related complications, and ABO incompatibility as a possible life-threatening condition was reported 97 times, leading to 4 deaths (
1 per 1,400,000 transfusions) and 29 cases of immediate major morbidity (1 per 200,000 transfusions) (10). By contrast, the SHOT initiative recorded only 19 cases of confirmed transfusion-transmitted infection (1 per 300,000 transfusions), including four cases of hepatitis B, three cases of hepatitis C, and one case of HIV. Eleven cases (one hepatitis A infection, one malaria, and nine bacteremias) were attributable to infections for which no testing of donations is performed (10). In the United States, the aggregated risk of transmission of HIV, human T-cell lymphotropic virus, hepatitis C virus, and hepatitis B virus is 1 per 34,000 blood units transfused (3), or less than half the risk of transfusion error (1 in 12,000 to 1 in 19,000 transfusions).
On the basis of current estimates of the risks of transfusion-transmitted infections (3) and transfusion errors (2,6,10), an anesthesiologist of a major general hospital or trauma center who transfuses an average 500 U of packed red cells per year can be estimated to transmit HIV infection once in 1000 years, hepatitis C once in 200 years, hepatitis B once in 120 years, and to administer blood to the wrong recipient once in 30 years, or once within his professional lifetime. In a university hospital with an annual volume of 20,000 blood transfusions or more, transfusion error is likely to occur once every year.
No data on transfusion errors are available from Germany, where the whole process of blood donation, preparation, and transfusion is subject to meticulous government regulations (11,12). These include requirements such as collecting the recipient’s blood for typing and crossmatching solely in prelabeled tubes and conscientious identification of both the recipient and the blood unit by the transfusing physician in persona. In addition, a pretransfusion bedside compatibility test is mandatory (11,12).
The incidents reported in the present paper were in violation of these regulations, and they highlight some specific areas in which failure is likely to occur: 1) labeling tubes after, rather than before, blood sampling the recipient’s blood, either by ignorance or negligence; 2) transfusing with no, or with insufficient, identification and bedside ABO compatibility test; and 3) similarity of patients by name or surgical procedure. All incidents occurred on the level of the transfusing physician rather than by failure of the blood bank. In three of our five cases, it was the bedside ABO compatibility test that prevented mistransfusion, whereas in the remaining two cases transfusion error was allowed to occur as a result of improper or neglected pretransfusion bedside compatibility testing.
Corrective actions taken in our institution to prevent transfusion errors in the future include the development of standard operating procedures for blood ordering and administration, periodic lectures for the personnel involved in these processes, and the implementation of an incident reporting system as part of a quality management program. Examples of incident reporting systems in transfusion medicine have been described in the literature (13,14).
In conclusion, our report agrees with previous studies showing that the greatest potential for further improvement of transfusion safety resides with the blood transfusion rather than with the manufacture of the blood unit. Proper performance of the blood transfusion is of paramount importance, as transfusion risk is shifting from blood bank errors and the transmission of blood-borne pathogens toward error on the part of the transfusionist (2,6,8). Increased awareness of this problem, in combination with a correctly performed and interpreted pretransfusion bedside compatibility test, may help to prevent avoidable transfusion errors that could result in fatal transfusion outcomes.
|
References |
|---|
|
TopAbstractCase 1DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. J. O'Connor, J. Hanson, and B. T. Finucane Induced hypotension with epidural/general anesthesia reduces transfusion in radical prostate surgery: [L'hypotension controlee, pendant l'anesthesie peridurale et generale combinee pour la prostatectomie radicale, reduit la necessite de transfusion]. Can J Anesth, September 1, 2006; 53(9): 873 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Solheim Universal pathogen-reduced plasma in elective open-heart surgery and liver resection. Clin. Med. Res., September 1, 2006; 4(3): 209 - 217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shander, D. Moskowitz, and T. S. Rijhwani The Safety and Efficacy of "Bloodless" Cardiac Surgery Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2005; 9(1): 53 - 63. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Marcucci, C. Madjdpour, and D. R. Spahn Allogeneic blood transfusions: benefit, risks and clinical indications in countries with a low or high human development index Br. Med. Bull., August 31, 2004; 70(1): 15 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Regan and C. Taylor Recent developments: Blood transfusion medicine BMJ, July 20, 2002; 325(7356): 143 - 147. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
