When the supernatant of a red cell unit shows visible discoloration suggesting hemolysis, should it be issued for transfusion?

A blood banker from India reports that sometimes a red cell unit (near its expiration) exhibits red discoloration in the supernatant. Apparently, the action taken upon discovering supernatant discoloration is subjective - some individuals issue such a product without hesitation, while others feel that the product is not within permissible limits to issue. The Indian blood banker is curious about what other colleagues use for an allowable limit of supernatant hemoglobin for red cell products. Does anyone routinely check each RBC unit prior to issue for supernatant discoloration, and if so, by what criteria are units considered acceptable or not acceptable?

The following replies were submitted in response to the above query.

1. According to John R. Hess, M.D., M.P.H., (attribution used with permission) who works for University of Maryland Medicine, it was reported at the FDA Pathogen Reduction Meeting on Aug 7-8, 2002 at the NIH that the U.S. product licensure standard is less than 1% hemolysis. Dr. Hess has special interests in blood product development. He was a speaker at the NIH meeting where he discussed Red Cell Viability Evaluation and Testing In Vivo. John is of the opinion that some degree of hemolysis occurs in all RBC units during storage. Dr. Hess reports that in Europe the standard is <0.8% hemolysis. To quote Dr. Hess, "Typically, SAGM or AS-1 storage systems show 0.4% hemolysis at the 6-week expiration date. Since a typical unit of RBC contains 50-75 grams of hemoglobin and about 160 ml of suspending plasma, anticoagulant, and additive solution, limiting mean (supernatant) hemoglobin concentrations for licensure are 300-450 mg/dL by U.S standards. Some individual donors whose RBC store poorly will have much higher hemoglobin concentrations and are allowed because average hemoglobin concentrations are lower. On the other hand, elevated plasma hemoglobin can be a sign of bacterial contamination or improper storage. Blood in bacterially contaminated bags is said to be dark because the bacteria consume the oxygen. With improper storage, such as inadvertent warming or purposeful microwaving, free hemoglobin should be more red. No good studies validate these observations. Dr. Hess adds, "What is needed is a sense of the range of normal RBC hemolysis and a reasonable suspicion about the potential for improper handling or bacterial contamination. The photograph below shows a blood bag with hemolysis that occurred during the course of leukofiltration.

Hemolysis of blood in the collection bag but not in the pilot tube after leukoreduction. Twenty-four other units drawn in the same study into the same lot of bags did not show hemolysis. Hb = 144 mg/dL.

Free hemoglobin, 144 mg/dl, was not present in the sample tube drawn at the same time as the bag. This degree of hemolysis is about half the lower limit of the legal standard and common in RBC storage studies, but worrisome. Perhaps others with long clinical experience might comment on the above example".

ADDENDA Nov. 20, 2002

2. A blood banker from the Pacific Northwest reports that at her facility they inspect each RBC unit before issue to check for hemolysis (as a surrogate of bacterial contamination), but they DO NOT spin down the product to assure that they have a cell-free supernatant to examine for color changes. According to this person, they can see hemolysis (when it is present) in the segment that was open to the component bag. Thus, in their hands, they do not need to centrifuge the unit to see hemolysis in the supernatant. In their facility it would be impractical and in fact impossible to centrifuge the unit prior to issue. What they can do for visual inspection at the time of issue is to inspect for hemolysis, note other visual signs of bacterial contamination i.e. a bloated bag due to production of gas and a black or dark discoloration of the bag. In a bacterially contaminated unit the supernatant in the tubing that opens into the bag will be hemolyzed. The supernatant in the distal segments will not be hemolyzed. For training purposes they contaminated an outdated unit and photographed it (see below).

This is one of the photos used for training at the Pacific Northwest blood banker's laboratories to demonstrate what a bacterially contaminated unit might look like. The supernatant in the tubing that opens directly into the blood container is grossly hemolyzed, compared with that of the more distal segments above.

(Editor's NOTE: Given that Dr. Hess reports that by U.S standards, at expiration, the allowable free hemoglobin in the supernatant can be as high as 450 mg/dL, one has to wonder how many RBC units show hemolysis sufficient to cause the unit supernatant to be brick red, yet NOT be due to bacterial contamination.)

3. The Editor is of the opinion that one of the issues at hand is that probably a large number of older RBC units have noticeable hemolysis, if one were to actually look for it using a cell-free supernatant. Assuming one looks for hemolysis in the supernatant of a centrifuged unit, what should be done with the unit that has up to the allowable 1% hemolysis at 42 days, since such a unit is likely to have a red supernatant coloration? Food for thought.

ADDENDA Oct. 28, 2004

4. The Editor thinks that the discussion in the article by Janatpour KA and colleagues in the July 2004 issue of Transfusion might be germane to this discussion.

ADDENDA May 30, 2007

5. A colleague asks "What is the scientific basis for the allowable 1% hemolysis standard for a unit of RBCs in the USA (or in the EU of 0.8%)? On what calculations were these values based?"

ADDENDA March 4, 2008

6. A colleague wonders what is the risk to a patient who is transfused with an RBC product that has had over 1% of the red cells hemolyze due to normal product storage, with release of free hemoglobin into the supernatant additive solution or plasma? According to Dr. John Hess who works for University of Maryland Medicine (attribution used with permission), hemolysis seems to have been a criteria for the duration of RBC storage ever since the original work of Rous and Turner in 1915, where it was essentially the only criteria available other than the rise in post-transfusion hematocrit before Ashby counts were described in 1919. Kendrick describes using hemolysis as a duration of storage criteria during World War II (p 224 in Kendrick DB. Blood Program in Wold War II, Office of the Surgeon General of the Army, Washington, 1964). Kendrick was describing whole blood stored in Alsevers' Solution (500 mL of each) so his criteria of less than 25 mg/dL of hemoglobin represented about 200 mg of free hemoglobin or a hemolysis of about 0.3%. He also notes that it was observed in Dec 1944 that a small number of units hemolyzed rapidly (page 552). The 1% hemolysis criteria for licensure was officially in place by the time of the CPDA-1 licensure study Zuck et al, Transfusion 1977) in 1977 and so presumably was worked out at the time of the licensure of CPD in 1957. It was used by Shields in the description of the original invention of CPDA-1 in 1968 (Shields C. Transfusion 1968). The 1% criteria seems to be a compromise between the desire for clear plasma and the above fact that some units hemolyze more without causing problems. The haptoglobin capacity to remove free hemoglobin would be capable of clearing about the free hemoglobin content of 40 units of RBC each with 1% hemolysis, so the actual amount of free hemoglobin should not be a problem with any modern storage solution. The 0.8% hemolysis standard used in Europe also seems to have no evidence-basis other than an attempt to "go us one better." In like manner, the recent suggestion by the FDA that the standard should be further defined with standard deviations and confidence intervals to improve blood quality is self-defeating in that it makes it very difficult to license new products.

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