A transfusion service physician in Missouri asked for help in investigating a report of suspect hemolytic transfusion reaction.

"This involved a 16 year-old Caucasian female undergoing an an orthopedical surgical procedure. She developed hemoglobinuria 2 hours after receiving two autologous RBC units that had been frozen and deglycerolized. A hemolysis check at the end of the processing of the units was said to be negative, but there was no documentation of a visual check of the units upon receipt at the hospital transfusion service (about 15 minutes away from the blood center).

The two RBC units were transfused because of a drop in Hb level from 11.8 g/dL before surgery to 8.6 g/dL during surgery, presumably due to blood loss. No adverse reactions were documented at the time of transfusion. The patient's Hb level at the time hemoglobinuria was observed was 10.6 g/dL; a Hb level had not been obtained immediately after transfusion or before the hemoglobinuria was observed.

Examination of the residual blood in the bags from the transfused units and their segments revealed gross hemolysis. At the time hemoglobinuria was observed, the recipient's plasma appeared grossly hemolyzed. A DAT test was negative. All post-transfusion clerical checks were correct. Osmolarity test performed on a segment was similar to the results obtained from checking a "normal" segment from another unit. All deglycerolizing fluids were re-checked and found to be correct. No tests for sickle cell trait were performed at the time the autologous units were collected.

The patient was given fluids and recovered quickly. The hemoglobin level remained stable on the day after the reaction, with no more blood transfused.

The two most likely factors causing hemolysis of the processed autologous units appear to be:

1. the cells were injured by the freezing/deglycerolizing procedure, and
2. something happened to the auto units after receipt and prior to transfusion (such as heating the units). I am unaware of anything unusual about the administration of the units.

I was wondering if anyone has encountered a similar situation and, if so, whether a cause was determined, and what the outcome was. Also, does anyone have any other ideas on what additional causes should be investigated? Since this patient has additional frozen units. I am hesitant to give them to her if she would need more blood. Does anyone have an opinion on what should be done with these units?"

The following responses have been received.

ADDENDA July 14, 2004

1. A colleague from Canada agrees with the premise that this phenomenon is likely due to a problem during the freeze/thaw or post-thaw handling. In order to help exclude an RBC membrane defect in the patient he recommends performing a Donath-Landsteiner test and an osmotic fragility test on a fresh sample of the patient's blood before thawing any of the other autologous units.

2. A Clinical Coordinator now in Florida recalls a similar case that occurred when she worked at a prestigious Chicago hospital in the 1970's where they used to store frozen autologous units for children scheduled for scoliosis surgery. She states they had performed a quantitative glycerol assay on the supernatant of the washed cells and found it to be very high. She suggests that the Missouri colleague's case could be attributed to inadequate glycerol removal, raising the possibility that personnel performing the deglycerolization had received inadequate training. She recommends that the deglycerolization procedure in use be thoroughly reviewed, and that they test the residual glycerol of some random supernatant samples as a quality control check, both by osmolarity and also by a quantitative method, to use as a comparison.

3. A Research Director at an academic institution in Northern California recalls a case in 1979 in which two units of in-vitro hemolyzed blood were heated in a microwave oven in the O.R. After the second unit was transfused, the nurse notice a small puncture in the bag through which blood was leaking very slowly. The residual blood in the bag was checked and found to be hemolyzed. The hole was possibly due to mechanical failure of the rotator in the microwave. The patient had frank hemoglobinemia and hemoglinuria which resolved uneventfully within several hours, with no medical intervention required. In this colleague's opinion, unless there is immune hemolysis, one would not expect to see any serious consequences from infusing thawed blood showing visible hemolysis at the time of infusion.

ADDENDA July 15, 2004

4. A Technical Support Program Director from a blood center in the Northeast U.S. offers the following points for consideration:

• Was a retention segment/sample retained by the laboratory? If so, how did the retention segment compare visually to the segments attached to the product after transfusion? If the retention segment was normal, i.e. no hemolysis, but the returned product/segment was hemolyzed, it would suggest some type of mis-handling of the products after release for the laboratory, e.g. storage outside the approved temperature conditions. This could include temperatures that were either too cold as well as too hot. What were the time frames from when the units were released/issued from the blood bank, the transfusions were administered and the hemoglobinuria was noted?
• Was a saline suspension prepared to perform the confirmatory ABO typing on the unit and/or compatibility testing? If the practice is to prepare a washed saline suspension prior to testing was any hemolysis noted at the time of washing the cells for testing? Preparing a saline suspension would be similar to performing a "simulated transfusion" as a check for adequate deglycerolization of the product prior to transfusion.
• The patient appeared to have the expected rise in hemoglobin level following the infusion of 2 units of blood. This suggests that intact red blood cells were present and only a small proportion of the red blood cells were hemolyzed.
• Were any solutions or medications added to the product at the time of administration, which may have not been isotonic? Since it was reported that the segments on the product following transfusion were also hemolyzed, this does not seem be a most likely explanation, but it could have been contributory.

Please submit comments to the e-Network Forum.

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