ADDENDA Jan. 8, 2003

14. From a pediatric hospital in a sunbelt state comes the following anecdote (verbatim) "This discussion on duplicate blood types comes at an appropriate time for a situation that happened in our hospital. (In a pediatric hospital, where sometimes the patient armband is on the patient's doll in their bed or on their IV pump connected to the patient.) A phlebotomist went into a teenage patient's room to collect a sample for a CBC (not a private room, two patients per room). The phlebotomist asked the patient in bed two if he was 'John Smith'. The patient replied 'Yes I am John Smith' and the phlebotomist, with name confirmation from the patient, proceeded to collect a blood sample. Upon leaving the room, the phlebotomist said thank you to the John Smith and for him to have a good day. The patient in bed one spoke up and said, 'but I am John Smith'. The teenage patient that had graciously placed his arm out and had his blood drawn started laughing. He was really Funny Randy and thought he would enjoy a practical joke." The sunbelt blood banker laments that this might be yet another reason for employing duplicate samples, when a previous ABO/Rh is not on file. Finally, the blood banker concluded that they were glad the sample in question was not for a type and cross!)

ADDENDA Jan. 9, 2003

15. A colleague who works for a major blood supplier in the US and who also performs transfusion laboratory assessments wrote the following "Great discussions on a very serious topic about requirements for second ABO/RH typing on first time recipients!! As an assessor, I can use these talking points as providing very applicable and helpful information for the Transfusion Services.

FOLLOW-UP ISSUE May 8, 2003 (A network dialogue on this issue.)

ADDENDA July 20, 2004

16. Editor's Note: This discussion may be of renewed interest in light of corrective action taken by a hospital in Florida following a recent transfusion-related fatality. Followup report - hospital now requires two typings (story links no longer available).

ADDENDA July 26, 2004

17. An immunohematology expert and blood bank scientist in Los Angeles wishes to point out data presented at the AABB meeting in 1982 by UCLA investigators (Rippee C, Morgan-Gannon S. Increasing detection of mislabeled specimens to reduce risk of mistransfusion. Transfusion 1982;22:461). These data showed how important it is to type a second sample from patients. Although not stated in the published abstract, the presentation showed an error rate (mislabeled 1st specimens) of 1 per 1,000 units transfused, or 1 for every 300 patients transfused. Eighty percent of these errors occurred when samples were not collected by specialized phlebotomists (e.g., in ER and surgery).

ADDENDA July 28, 2004

18. A transfusion medicine physician in Detroit expresses the opinion that separately collected specimens for determination of a patient's ABO group should be mandated by AABB Standards as well as by the CAP accreditation program. Her institution had been retyping a single sample by different technologists. They had been considering a mandated second sample, but were reluctant to implement that approach until they had two ABO transfusion errors occur due to mislabeling of samples on patients with no prior blood types on file. To avoid additional such cases, they will soon be mandating a second (in their case, fingerstick) sample for patients who do not initially type as group O, and have no previous ABO records in their LIS. Patients requiring emergency group O RBCs in the ER and uncrossmatched group specific blood (usually in surgery), will by exempt. The responding physician has advised surgeons performing elective procedures that they should order an ABO testing of their patients during pre-surgical clinic visits, if they wish their patients to avoid being stuck twice on the morning of surgery.

ADDENDA August 3, 2004

19. A colleague in Austin, Texas reports that her institution currently requires a second type to be performed on the same sample but from a fresh cell suspension.  This is accomplished by using cells directly aspirated from the same clotted sample, may be performed by the same technologist, and can be run in a batch. She adds that as she has read through the responses posted above, and based on reading about recent events of transfusion-related deaths in the US, she has changed her views and agrees with repeat typing on two separately collected blood samples.  She agrees with the respondents who state that the second sample could be from another department drawn at a different time.  She thinks that if a second sample was not available in the lab that a finger stick sample would be acceptable to recheck the forward type.

ADDENDA August 10, 2004

20. A colleague in North Dakota reports that her facility has found the information and comments concerning performance of ABO/Rh testing at least twice as part of pre-transfusion testing to be quite interesting and helpful. Currently, her hospital is in the process of implementing a plan that requires concordance of two ABO/Rh types based on testing two separate samples for all patients before transfusion. If the patient's ABO/Rh is in their database, that typing result can serve as one of the two required types for comparison. One question that came up with this plan concerned pediatric patients. Do those facilities that require concordance of two ABO/Rh types done on two separate samples require this practice be applied for pediatric patients before blood is infused? If a facility requires concordance with a second ABO/Rh for pediatric patients, what is the process and does this process vary from adult patients?

ADDENDA August 15, 2004

21. A transfusion medicine colleague completely agrees that determining a patient's ABO/Rh based on typing specimens from different phlebotomies is a safer practice than relying on a single blood typing result.  However, there are huge staffing implications and in her facilities the phlebotomists are not plentiful.  She is investigating systems that read bar codes on the patient's armbands and print out specimen labels at bedside to reduce chances of identification errors.  These are expensive systems but the additional safety factors should considered in the hospital setting at budget time. However, she correctly points out that if the wrong armband is placed on a patient (it happens), this won't prevent a specimen labeling error, even if more than one sample is tested for ABO/Rh.

ADDENDA August 24, 2004

22. A transfusion service supervisor in Northern California reports that four years ago her hospital instituted a policy requiring that a patient should be tested for ABO on two separate sample to verify their ABO before initiating transfusion therapy. This policy was instituted as a corrective action following what was described as 'a very close call'. At first, the nurses drawing pretransfusion specimens had a little difficulty understanding the reason for this policy, but after a few months, compliance with the policy was achieved. In response to the colleague from North Dakota (posting #20 above), they do include children and neonates in their policy, if those small patients are receiving red cell products from non-group O donors. In general, they use pediatric aliquots from group O Rh negative donors and pediatric aliquots of FFP from group AB donors, but occasionally they cannot get a group AB, CMV negative platelet product. In those cases, they verify the patient's ABO using a new blood specimen, or if available they use a cord blood sample, or obtain a specimen from their Hematology department. Her question is for colleagues who follow a similar policy, 'how are you handling your trauma cases?' Their ER situation is that there is not a person available to independently identify the 'Doe' patient and draw a second specimen. Are group O Rh negative donor units being used for these trauma patients even if you have one 'good' specimen? Is a second specimen required before issuing group specific non-group O donor units?

ADDENDA August 26, 2004

23. A Texas colleague reports that her institution is part of a system of 5 hospitals, one of which is a children's hospital. They require a second sample for blood typing on all transfusion patients - adult, pediatric and trauma. If for some reason they cannot get a second sample for typing, the physician must sign an emergency release authorization form and only group O RBCs are issued. The second sample for typing may come from:

• Historical information from any one of their sister hospitals (they have access to each other's database)
• A sample already in the laboratory, drawn at a different time
• A fingerstick sample collected and presented to the blood bank at the time that nursing staff comes down to pick up a unit. The ABO/Rh of the fingerstick sample is done by one tech while another tech does the paperwork and checking of the unit.

The unit is NOT released to the nursing staff until the type check is completed. The blood bank tech notes in the testing record whether or not a "QTC" (quick type check) sample is required, so nursing knows before they send for any blood products.

For several reasons, she reports that they require that a "QTC" sample be brought to blood bank for ALL surgical patients, regardless of whether or not they have a second blood typing result already on file. It is required that this sample be collected in OR immediately prior to surgery (either in OR holding or in the surgical room). It is usually collected as the IV line is placed. This sample is also brought to the blood bank when the nursing staff comes down to pick up blood. When this colleague first started at this hospital, she considered this protocol to be "overkill", but they have had numerous instances where OR staff comes to pick up blood and the QTC blood type does not match the pre-surgery sample. It usually turns out that the chart that nursing has in the OR room belongs either to the patient that was operated on earlier or to the patient in the next room.

She adds: "If the discrepancy had not been caught, all samples sent to the lab would have been labelled with the wrong patient information since they use pre-printed labels from the chart to label interoperative samples. It can also be very difficult to recognize a transfusion reaction in a sedated surgical patient due to the dynamic situation in surgery. This provides an extra level of security since the first sample is usually collected by our pre-op admissions office (for outpatients) or on the floor by a phlebotomist, and the second sample is collected after the patient has been transferred into the surgical area."

ADDENDA Dec 28, 2004

24. A colleague in Florida reports that earlier this month (December 2004) they implemented a policy that required confirmatory ABO/Rh typing for non-group O patients for whom they do not have a historical blood group. Initially implemented at their healthcare network's main hospital, they will be extending a confirmatory ABO/Rh typing protocol to two smaller community hospitals after January 1, 2005, and a new hospital will open with this procedure in place. The Florida colleague has graciously provided a copy of their protocol.

The previous discussions on this topic in the CBBS e-Network Forum were extremely helpful - both for focusing his own thoughts on the issue, and for convincing his administration, nursing staff (and even some blood bankers) of its importance. Transfusion medicine is only a small part of his practice, and he has found the forum to be a valuable resource in keeping abreast of important issues and developments in the field.

ADDENDA Dec 31, 2004

25. A colleague in Hungary reports that their Transfusion Regulations demand that if the patient has no record of his/her blood group, the AB0 group and the Rh type should be done twice from the same sample. The clinicians have to check the blood group of the patient at the bedside, either on a card or with liquid reagents on a slide. This can be called the second, independent test. Most hospitals do not have their own pretransfusion laboratories and send their samples to a central blood bank laboratory. The responding Hungarian's former working place serviced more than 30 hospitals, among them the National Trauma Center, from whom they received daily 20-30 samples in batches of 3-8, all of them, always emergent, impossible to require a second sample before releasing blood for transfusion.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Assistant Editor & Moderator