A blood banker in Virginia wants to know if members of the e-network forum have developed policies defining when platelets and RBCs should be cultured as part of a transfusion reaction investigation. He wonders if it is possible for a standardized policy to emerge given the widely varying thresholds in the past for culturing transfused blood products. For example, Ness et al. (TRANSFUSION 2001;41:857) report an incidence of septic reaction to platelet transfusion of 1 in 15,000 products (whether whole blood-derived or apheresis). Heddle and Kelton in Popovsky (ed.), TRANSFUSION REACTIONS, 2nd ed., page 51 report a frequency of FNHTR following platelet transfusion to be 11-37%. Thus, the inquiring blood banker is concerned if one were to culture implicated products for all transfusion reactions, one would be embarking on a costly endeavor with only occasional benefit. At present, the inquiring blood banker's facility cultures blood products implicated in a transfusion reaction on an individual basis after evaluating the patient. No particular guideline or algorithm is followed. Perhaps the input from the e-network forum will assist the inquiring member as well as others.

The following replies were submitted in response to the above:

1. A pediatric blood banker reported that at her facility they have a protocol to culture aerobically RBCs, Platelets or both based on which product was administered at the time a patient's temperature increases more than 2F, in consultation with the Transfusion Service Director. Many of the febrile reactions are called for patients who are neutropenic and who are receiving (or have recently discontinued) broad spectrum antibiotics. They request that a culture be taken from the child using a site different from the one in which the blood or platelets were infused. This brings up another interesting question which is how often is bacterial growth in a culture supressed because a blood culture was taken from a line through which antibiotics were being administered.

2. Another blood banker commented that her facility is part of an alliance of 5 hospitals. Their policy states that any platelet product inplicated in a transfusion reaction will be cultured. RBC's are cultured if one of the patient's symptoms is fever and the temperature increased by 2C or more.

3. A blood banker who works in Bethesda, Maryland (at the AABB National Headquarters) suggested that the e-network forum consider referring to the BaCon Study guidelines for what constitutes a possible bacterial contamination and what should be cultured. Although that study has ended, these guidelines are still posted at the CDC web site. Study Coordinator Matt Kuehnert MD at the CDC can be contacted for information Phone: 404-639-6413).

4. A California blood banker said that their current transfusion reaction workup for components requires culture of the unit on all reactions where there has been an increase in temperature of >2 degrees F or 1 degree C and the ending temperature is >100.5F

5. Another blood banker said that the policy for their system (of 4 hospitals) is that if there is 2 degree C raise in temperature or a drop in blood pressure, they will culture the unit. In their last 20+ years with the system they had two cultures come back with positive results (Staph species attributed to contamination as patient blood cultures were negative).

6. Yet another blood banker said that for what it is worth, their policy is to perform a culture for a rise in temperature > 2 C, symptoms that include hypotension, shock or death, or at the request of the physician, either the reporting physician or the blood bank physician. This applies to red cells and platelets equally. They have had positive cultures both in the context of clinicallly obvious sepsis reactions, and occasionally when the only reason was temp rise > 2 C.

7. A blood banker in Central California says that his blood center has never had a strictly formal policy determining when they should initiate bacteriologic studies as part of the transfusion reaction investigations that they perform (or consult on). However, their basic approach has been to initiate a gram stain and aerobic/anaerobic cultures of the patient's blood, a representative sample from the implicated component (from WITHIN THE BAG - not within the segments - when at all possible), and IV solutions within the administration tubing when any of the following conditions develops within 90 minutes of a transfusion:

1. (i) Temperature > 39 degrees C or > 102 degrees F,* or (ii) Temperature rise of > 2 degrees C; or > 3.5 degrees F; (*Note: "a" is somewhat more subjective. For example, if the patient's pre-and post-transfusion temperatures were 38.7 and 39.1 degrees C, respectively (and no likely manifestations of a septic transfusion reaction existed), they would not initiate a septic transfusion reaction workup; however, if the pre- and post-transfusion temperatures were, say 37.3 and 39.1 degrees C, respectively, they would initiate the workup, even if no other suspicious findings existed);
2. Severe, sustained rigors;
3. Elevation in heart rate of > 40 BPM;
4. Reduction or elevation in systolic BP of > 30 mm Hg;
5. Nausea, vomiting, dyspnea, and/or lumbar pain, when not readily explainable by another likely cause;
6. Hemoglobinuria, DIC, and/or renal failure (if these problems developed, I would recommend initiating sample acquisition for bacteriologic studies--and also ask that the gram staining actually be started--even while the workup for an acute hemolytic transfusion reaction is in progress);
7. Visible signs of significant clotting, hemolysis, or abnormal color (e.g., brownish, opaque, muddy, or purple) observed within the implicated unit after the transfusion has been initiated.One final comment: When they suspect a septic transfusion reaction, they always remind the laboratory physician/technologist coordinating the investigation to perform the aforementioned cultures at refrigerated (for cryophilic organisms such as Yersinia enterocolitica), room, and body temperatures for RBC/plasma components, but only room and body temperatures for platelet components.

ADDENDUM Oct. 8, 2001

8. A blood banker from North Carolina commented that her hospital's policy is to culture all platelet products involved in reported reactions and red cell products that as a result of being manipulated, have become open systems, i.e., washed, deglycerolized. Additionally, other products may also undergo culture, if the culturing is requested by a pathology resident or other laboratory physician, as a result of a transfusion reaction investigation. The e-network forum is referred to: S73-040A in Transfusion's AABB Annual Meeting 2001 Abstract Supplement entitled "Culture results on platelets involved in transfusion reactions: One institution's 7-year experience".

9. Another blood banker commented that his facility gives approximately 11,000 platelet transfusions a year. He thinks it is essential and their standard practice is that each transfusion reaction with at least a 2 degree F rise has a gram stain and culture done (this is in their blood bank SOP). In addition each reaction reported to the blood bank is reported immediately, 24 hours a day, for evaluation by the blood bank physician (usually after the DAT is done); if the symptoms are severe but without a fever immediately posttransfusion (e.g., severe pulmonary difficulty, bed-shaking chills or acute hypotension) the physician asks that a gram stain and culture be done.

ADDENDA Mar. 28, 2006

10. Editor's note: It has been over four years since the last posting to this discussion. A colleague in New York reactivates this discussion by reporting that the hospital network for which he works is interested in standardizing the work up of transfusion reactions, specifically when to culture an RBC unit that has been implicated in a transfusion reaction. He is aware that some hospitals culture each RBC unit involved in a suspected transfusion reaction, while other hospitals culture an RBC unit only if the patient experiences at least a 2ºC rise in temperature while being transfused. He also believes that some hospitals perform a Gram Stain on a suspect RBC unit and culture the unit only if the staining result is positive for bacteria. The inquiring colleague wonders what current protocols are being used by others institutions to cause RBC units to be cultured.

ADDENDA June 26, 2007

11. The Assistant Director of Transfusion Medicine and Cellular Therapy at an academic center in New York reports that her microbiology laboratory has a question about how others obtain a specimen for Gram stain and culture from a segment of a red cell unit that is implicated in a possible septic reaction. She acknowledges that they do not normally sample segments for microbiologic studies, but they have had a few cases lately where a transfusion reaction was suspected more than a day after a transfusion occurred, and the implicated product containers and tubing were already discarded. She was wondering if any blood banks send segments for culture, if the clinicians are highly suspicious of bacterial contamination of a red cell unit (and the returned unit container is not available). Second, if segments are being sent to the laboratory, how does microbiology handle them? The microbiology staff were trying to sample segments using a sterile needle, but one of their technologists recently suffered a needle stick injury because of the difficulty of this procedure. The microbiology lab is now refusing to culture any segments until a better (safer) sampling procedure is developed. The inquiring colleague would appreciate feedback.

ADDENDA July 2, 2007

12. A transfusion medicine physician in Cleveland reports that regarding the question of culturing 'preexisting' unit segments - i.e. those made at the time the unit was collected - in her opinion these are virtually worthless as far as assessing for bacterial contamination of the unit itself. The reason is that the initial entry of bacteria is at such a low level (perhaps 10 or even less microorganisms per unit) that the chance that even a single organism will wind up in a pre-existing segment is remote. She remembers only one out of about 50 instances with bacterially contaminated platelets that such a segment was thought to be positive. She guesses there is no harm in culturing this, realizing the yield will be incredibly low, except for the possibility of false positive cultures. In handling such a segment, her lab allows the sample to flow into a sterile tube and then the sample is removed from the tube - this eliminates the chance of an accidental needle stick. At least with platelets, her facility refrigerates temporarily the 'empty' bags after pooling in the event a transfusion reaction is reported. That way the actual bag contents could be sampled. This would not readily apply to RBC units - which have an extremely low rate of bacterial contamination anyway.

ADDENDA July 16, 2007

13. A transfusion medicine physician in Houston reports that in her experience the likelihood of recovering bacteria from a segment attached to a possibly contaminated red blood cell or platelet unit is nearly zero. In addition, the possibility of harm (namely, a needlestick injury as was described in the question) is high. If the suspect product bag itself is not available for direct sampling for culture, then there is very little possibility of recovering micro-organisms. She would strongly recommend that blood/component administration procedures be revised to state that a component bag and attached blood administration set be saved and sent to the transfusion service, if there is any suspected reaction. If a blood product unit is truly contaminated, the recipient will likely (although not always) become symptomatic either during or immediately after the transfusion. In many instances the symptoms are not subtle.

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