Policies and practices regarding transfusion of febrile patients

A transfusion medicine physician in the heartland of America reports that her hospital's practice is to defer initiating a blood product transfusion, if a patient is febrile. She realizes that this approach could delay medically needed transfusion therapy, and they are looking for evidence to update their internal policies so that febrile patients that need components may be transfused. What is the policy and practice at other institutions regarding transfusion of febrile patients, and upon what evidence are these policies/practices based?

The following comments have been received.

ADDENDA Dec. 6, 2005

1. A transfusion medicine physician in North Carolina reports that the major concern in transfusing a febrile patient is that a hemolytic reaction, which is often detected due to a rise in temperature, may be masked in such patients. Alternatively, unnecessary workups for a hemolytic reaction may be triggered in a patient having frequent fever spikes. He acknowledges that it is often prudent to postpone a transfusion until a patient's fever is brought under control. However, he cautions that if the patient's medical condition warrants immediate transfusion, the transfusion should not be delayed.

ADDENDA Dec. 7, 2005

2. A colleague in Chicago reports that her hospital's Blood Transfusion Policy / Procedure does NOT preclude administration of a unit of blood to a febrile patient. Rather, the transfusion of a febrile patient is considered to be a medical decision. She adds that over the years she has provided transfusion education to their nurses, many of whom inform her that during nursing school they were taught that if a patient is febrile prior to transfusion, the ordering physician should be called to determine if the unit should be transfused or wait until the patient's temperature becomes "normal". In some cases the physician might order the transfusion to be given right away regardless of the fever. In other cases the physician might order antipyretics to lower the patient's fever prior to initiating the transfusion. The physician's decision is based on the medical presentation and the patient's clinical situation. A quick review of the AABB Primer of Blood Administration (rev. January 2004) does not indicate that transfusion should be withheld if a patient is febrile.
   
   Editor's Note: The AABB Primer was updated in May 2005 (AABB membership required to access) and states in Chapter 5, page 7 "fever..1) may be cause for delaying the transfusion 2) could mask a symptom of an acute transfusion reaction and 3) may compromise the efficacy of platelet transfusions."
   
   
3. A transfusion medicine physician in Los Angeles reports that in his practice febrile patients often need transfusion, and that a pre-existing fever may mask a febrile non-hemolytic transfusion reaction. However, in his experience, such reactions are generally not clinically serious. He acknowledges that a problem can arise when a febrile transfusion reaction investigation is initiated for a patient who is already febrile before a transfusion began. As a diagnosis of exclusion, it is difficult for him to diagnose a febrile non-hemolytic transfusion reaction for a pre-existing febrile patient, or for a patient who has reason to be febrile (an infection or atelectesis for example). His transfusion service has no policy to preclude the transfusion of a febrile patient, which occasionally leads to extra transfusion reaction workups.
   
   
4. Dr. Jim AuBuchon at Dartmouth (attribution used with permission) reports that while fever can be an important sign of a transfusion reaction, there is no need to delay a transfusion because of a pre-existing fever. He adds that while there are some data to suggest that increased body temperature may shorten the lifespan of platelets, the data are not absolute on this point, and the effect as not absolute, anyway. He comments that if a transfusion is medically indicated, it should be given! His institution has been using this approach for many years (> 12!) without difficulty. They have defined a febrile reaction both in terms of an absolute temperature being reached (38.0C) in addition to a rise in temperature (of at least 1C).

ADDENDA Dec. 8, 2005

5. An experienced transfusion medicine physician in the Midwest reports that her personal opinion on this matter is that with all things involving patient care, common sense and attention to the patient's needs prevail. There are at least two reasons to consider withholding transfusion until the patient's fever is adequately managed, 1) it makes it more difficult to detect adverse reactions to the transfusion and, 2) there is some old literature that fever may reduce red cell survival (Mollison, sixth edition). She acknowledges that it is the former concern that drives policies to mandate withholding transfusion until the patient's temperature is resolved. Such a policy is logical and well-founded providing that it allows for a physician to make the final decision regarding whether the transfusion should proceed or not. Policies should then also address how the monitoring of an already febrile patient may need to be modified to improve the chances of detecting an adverse event.

ADDENDA Dec. 13, 2005

6. Colleagues in Connecticut report that their academic center does not consider pre-existing fever as a contraindication for blood transfusion.  If a transfusion is deemed necessary by the clinical team based on the clinical situation, transfusion proceeds, and the usual transfusion monitoring is exercised. If a patient develops a rise in temperature of at least 1C or 2F above baseline, or if other complaints develop that fall under their routine requirements for a transfusion reaction work-up, the transfusion is stopped and a complete reaction work-up is initiated.  This has been their practice for years (>20 years).
   
   The responding colleagues provide the following data from their center: During the time period 1/1/2005 to 12/7/2005, a total of 86 patients were evaluated for a transfusion reaction due to an increase developing in the patient's temperature during or after a transfusion. Of these patients, 23% were already febrile immediately before the transfusion, whereas 35% were NOT febrile immediately before the transfusion but had been febrile hours or more before the transfusion. In many cases the febrile patients had been given an antipyretic to lower their temperature.  None of these patients (23% + 35%) experienced a hemolytic transfusion reaction. The responding colleagues acknowledge that their center may do unnecessary transfusion reaction work-ups for patients who manifest fever prior to transfusion, since they cannot distinguish between a febrile transfusion reaction, a hemolytic reaction, or fever related to a patient's underlying condition, without doing a work-up. They also acknowledge that even if a transfusion was delayed while waiting for a patient's fever to subside (following administration of antipyretics or by merely waiting), it is possible for the patient's fever to return in temporal association with the transfusion, even if the cause of the fever is not the transfusion.

Submit comments to the e-Network Forum at enetworkforum@cbbsweb.org

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

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

Elizabeth M. St. Lezin, MD
CBBS e-Network Forum Associate Editor & Moderator