A colleague requests written criteria for nursing staff (or other healthcare providers who administer blood products) to initiate a transfusion reaction work up, based on changes in blood pressure. Her request is based on a policy at a hospital in a North Atlantic state where it is required that a transfusion reaction work up to be initiated when a patient experiences an acute change in blood pressure (either up or down) of at least 30 mm Hg in association with a transfusion. The nursing staff at that hospital reportedly feel that the aforementioned criterion is arbitrary and is too stringent. They comment that other institutions have transfusion reaction criteria that either make no mention of blood pressure changes, or merely make mention of generic statements about the development of acute hypotension or hypertension.

The following comments have been received.

ADDENDA Mar. 10, 2006

1. A colleague at a 1000+ bed hospital in the Midwest comments that in their study entitled "Unreported Transfusion Reactions in A Very Large Community Academic Teaching Hospital" which appears in Transfusion 45:3S, Sept 2005, only 5% of events meeting the hospital's criteria of a possible transfusion reaction were reported by nursing. Stringent retrospective review of transfusion records (n=16,606) produced 1,703 "unrecognized reactions" which had not been reported to the blood bank. Follow-up of the patients' charts did not reveal any "significant" unreported "reactions". Most of the unreported "reactions" during transfusion were limited to blood pressure changes of at least 30 mm Hg (either an increase or decrease).

2. A colleague in California reports that at her hospital, they are also addressing the criteria to use for a suspected acute transfusion reaction, based on changes in blood pressures. They perform audits of 'clinical records' pertaining to transfusions, and look for evidence of "hang times" exceeding four hours, and "febrile and hyper/hypotensive reactions" that occur in association with transfusion, but which are not reported to the lab. They have found several examples of what appear to be 'unreported' reactions. They have extreme difficulty in convincing nurses that blood pressure changes must be reported as possible transfusion reactions. At their hospital, they have used a change (up or down) of 40 mm Hg in either the systolic or diastolic pressure. However, they are considering revising the criteria to a more universally understood nursing standard of a 20% change over baseline.

ADDENDA Mar. 12, 2006

3. A medical technologist at a Regional Health Center in Texas reports that in their 350 bed hospital they use a DROP in the systolic blood pressure of more than 20mm Hg and a DROP in the diastolic blood pressure of more than 10 mm Hg as criteria of a possible "severe symptom observation". They use an increase in the blood pressure, without any specific number, as a criterion of a "less severe symptom observation". She acknowledges that at the time they set this criteria, they could NOT find compelling literature (evidence) to support their use of the aforementioned criteria. All they could find were generalities. One of their Pathologists decided arbitrarily what changes in blood pressure they would put on their Transfusion Reaction form! See is "happy" to see this problem addressed by the e-Network Forum, as she feels that their local criteria may be a bit stringent. She has graciously provided an example of the 'two-sided' form (pdf file) that their nurses print from a ward computer, when a transfusion reaction is suspected. They are told during orientation and competency fairs that the treatments on the back of the form are suggestions they might offer, if the physician asks for their opinion.

ADDENDA Mar. 14, 2006

4. A physician in Texas reminds us that the Bacterial Contamination of Blood (BaCon) Study collected data from January 1, 1998 through December 31, 2000, and used a drop in systolic blood pressure > 30mm Hg (from pre-transfusion values) or a rise in systolic blood pressure > 30mm Hg as a sign of possible bacterial contamination. (See Identification of transfusion reaction resulting from bacterial contamination).

5. A colleague at a Georgia medical center comments that the use of an arbitrary change in a patient's blood pressure (up or down) in association with transfusion could trigger an inappropriate transfusion reaction work-up.  For example, if a patient was in shock with a blood pressure of 80/50, but improved with transfusions to a blood pressure of 120/75, under a 'rule' where a change in blood pressure (up or down) of more than 20% triggers a reaction work up, such an improvement would need to be investigated as a possible transfusion reaction! He suggests that changes in blood pressure by themselves (especially modest changes) should be accompanied by one or more other symptoms suggestive of a possible transfusion reaction, before a reaction work up is triggered.  He cautions that "Much careful consideration/discussion should take place before making this a part of the reporting process for suspected transfusion reactions."