Is This a Hemolytic Transfusion Reaction?

Christopher M. Lough, MD; Leonard I. Boral, MD, MBA; Julie Ribes, MD


December 11, 2013

In This Article

Differential Diagnosis of Nonimmune Hemolysis

Physical/chemical causes. These causes of hemolysis are most often secondary to improper handling of blood products. Reasons include inadvertent heating or freezing with subsequent thawing of the red blood cell unit; exposing the donor cells to a hypotonic solution, such as D5W; and rapid transfusion of the red cells through a small needle (21 gauge or less) while using a pressure cuff around the red cell bag, resulting in mechanical hemolysis. A difficult phlebotomy might also result in hemolysis, and the pink to red plasma in the test tube might lead to the mistaken conclusion that in vivo hemolysis has occurred.

In vivo causes of physical hemolysis include red cell shearing from mechanical heart valves; damage from severe burns; and repeated soft-tissue trauma, such as "foot strike hemolysis" from long-distance running. In our patient, hemolysis detected in the first specimen was suspected to have resulted from a traumatic venipuncture, but the repeat specimen also demonstrated hemolysis and clotting, suggesting that something beyond poor specimen collection was occurring.

Hereditary and acquired red cell defects. Hereditary and acquired alterations of red cells causing hemolysis can be divided into defects of the red cell membrane, the hemoglobin molecule, and red cell enzymes. When the red cell membrane is misshapen or weakened, as is the case with hereditary spherocytosis/elliptocytosis and sickle cell disease, hemolysis of the patient's own red cells can be precipitated by various stressors, such as infection.

In paroxysmal nocturnal hemoglobinuria, an acquired red cell gene abnormality results in the deficiency of red cell glycophosphatidylinositol, thereby making the red cells more susceptible to hemolysis by complement. Thalassemias, including alpha- and beta-thalassemia, can cause abnormalities in red blood cells (eg, Heinz bodies), which are taken up by the spleen and destroyed.

The prototypical enzyme defect of red cells is glucose-6-phosphate dehydrogenase (G6PD) deficiency. In G6PD deficiency, oxidative stress, as well as certain foods (fava beans) and medications (quinine), can deplete the body's stores of glutathione, causing damage to the red cell membrane, which results in clearance of the defective cells by the spleen.

All patients with hemolysis are potentially hypercoagulable and susceptible to thrombosis, with increased frequency of deep vein thrombosis, PE, cerebral strokes, and myocardial infarctions. The mechanism is related to the binding of nitric oxide by free hemoglobin. Nitric oxide is released by endothelial cells to prevent platelet aggregation and to keep blood vessels relaxed. When nitric oxide is removed, there is a tendency for the opposite to occur, with resulting enhanced platelet aggregation and vasoconstriction.[3]

Our patient had no history of any hereditary or acquired red cell disorder, and the acuity of his presentation would argue against any long-standing disorders such as these.

Thrombotic microangiopathic hemolytic anemia. T-MAHA is a set of disorders that includes hemolytic uremic syndrome (HUS), atypical HUS, malignant hypertension, thrombotic thrombocytopenic purpura (TTP), and disseminated intravascular coagulation (DIC). In this category, hemolysis is secondary to microthrombi occurring throughout the microvasculature, resulting in lysis of red cells as they attempt to pass through small and obstructed blood vessels. Schistocytes on peripheral blood smear are an important diagnostic feature of these disorders.

In HUS, the precipitating factor for thrombi formation is typically infection with Escherichia coli O157:H7, which creates toxins that cause microthrombi throughout the microvasculature. Atypical HUS is a relatively newly recognized condition involving an acquired genetic alteration that causes defective anti-complement factors (factor H, factor I), leading to increased complement activity and a T-MAHA state.[4]

TTP is caused by an autoantibody to ADAMTS13. The resulting lack of the ADAMTS13 metalloenzyme, normally found in the plasma, means that this enzyme is not available to break down high-molecular-weight von Willebrand factor to its normal moderate-sized state. If it is not broken down by ADAMTS13, high-molecular-weight von Willebrand factor, causes platelets to be abnormally activated, and they aggregate with only mild shear stress throughout the vasculature.

Other non-TTP conditions that can sometimes cause a T-MAHA picture with microangiopathic hemolytic anemia (schistocytes) and thrombocytopenia include metastatic cancer, bone marrow transplant, pregnancy, HIV, certain medications (cyclosporine, tacrolimus, ticlopidine, clopidogrel), and DIC.

DIC was high on the differential diagnosis list in this case. The patient's clinical picture of possible PE with a subsequent decrease in hematocrit and platelets would fit with DIC. As small blood clots form within the blood vessels, platelets and coagulation proteins are consumed and inappropriate bleeding occurs, as in our case. The lack of schistocytes on the peripheral smear goes against the diagnoses in this category, but because of the difficulty in collecting acceptable blood specimens in our patient, confirmatory testing (prothrombin time and activated partial thromboplastin time) could not be performed.

Also ruled out were a transfused donor red blood cell unit contaminated and hemolyzed by gram-negative organisms causing gram-negative shock and DIC, and a snake bite resulting in DIC.

DAT-negative autoimmune hemolytic anemia due to IgA. An immune DAT-negative hemolytic anemia is possible, but very unlikely. A send-out test demonstrating IgA coating the patient's red cells would have to be done to prove such a condition. Remember that the DAT reagent only contains anti-IgG and anti-C3bd. It does not contain anti-IgA, anti-IgM, anti-IgD, or anti-IgE.

Infection. The last category to consider is infection. The patient had never traveled to a malaria-endemic area, so overwhelming malarial infection was unlikely. There was no evidence of bartonellosis or babesiosis on the blood smears. Haemophilus influenzae type b has been shown to cause hemolysis by way of polyribosyl ribitol phosphate (PRP) binding to erythrocytes and activating complement when both PRP and its antibody are present.[5] However, this is rarely life-threatening, and does not occur with the rapidity of the hemolysis in this case.

The patient had unfortunately died before sufficient microbiological work-up could be completed, but the following were found on further review. The Wright-Giemsa-stained blood smear demonstrated scattered, large, rod-shaped bacteria and a paucity of red blood cells (Figure 4). The blood culture plate from the complete blood count specimen grew gram-variable rods (Figure 5) and a double zone of hemolysis (Figure 6). The organism was eventually identified biochemically as Clostridium perfringens. Of note, the segments from the red blood cell donor units transfused into this patient showed no hemolysis upon examination at the time of the transfusion reaction work-up.

Figure 4.

Wright-Giemsa stain of peripheral blood smear. Note the reduction in red blood cells and the large bacterial rods.

Figure 5.

Gram stain of blood culture specimen colony growth, showing sheets of gram-variable, boxcar-shaped rods.

Figure 6.

Sheep-blood agar plate from patient's blood specimen. A colony showing a double zone of hemolysis is seen.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.