FINAL DIAGNOSIS: TRANSFUSION ASSOCIATED CIRCULATORY OVERLOAD (TACO)
When a patient receives a transfusion and then develops dyspnea, several possibilities are in the differential diagnosis: acute intravascular hemolysis, bacterial contamination, anaphylactic transfusion reactions, TRALI, and TACO. TACO is a relatively common but under-diagnosed complication of transfusion. It is also a more common complication of transfusion than TRALI (Fiebig, et al). Simply put, TACO occurs when an increased intravascular volume with elevated hydrostatic pressure exceeds the physiologic capacity of the cardiovascular system resulting in pulmonary edema.
Classically, patients with TACO have symptoms within 2 hours of the transfusion and may also present with acute elevations in blood pressure and widened pulse pressure, elevated pulmonary wedge pressure, tachycardia, and other signs of fluid overload. The consequences of TACO may be the need for mechanical ventilation, prolonged hospitalization, and increased morbidity and mortality. The incidence of TACO has been reported to be as low as 1% (Popovsky, et al) and as high as 8% (Bierbaum, et al). TRALI, another important diagnosis in the differential, has been found to be one of the most frequent causes of death following transfusion, with mortality rates reported as high as 10% (Tobian, et al).
In contrast to TACO, patients with TRALI often are hypotensive and have low-to-normal pulmonary wedge pressures. TRALI has been theorized to occur from lung injury associated with increased pulmonary vascular permeability secondary to donor leukocyte antibodies or infusion of neutrophil priming agents (e.g. lipids, CD40 ligand). A subset, approximately, 15% of TRALI patients have been found to be hypertensive at the onset of symptoms (Andrzejewski, et al). The natural progression of illness in TRALI is also similar to TACO - often occurring within two hours of transfusion (Tobian, et al).
Differentiating TACO from TRALI is complicated by the fact that the two entities are not mutually exclusive - that is to say, a patient can have elements of both complications. The treatments in these two conditions are different and improper treatment can lead to increased morbidity and possibly mortality of patients. Thus, it is important to make the correct diagnosis as early as possible. An objective, reliable biological marker to help distinguish TACO from TRALI would be an extremely useful tool for a Transfusion Medicine physician to have but is not currently available. It remains a clinical determination.
A recent paper by the University of Michigan Transfusion Medicine group (Zhou, et al) evaluated the utility of immunoassay measurement of B-natriuretic peptide, a polypeptide secreted from cardiac ventricles in response to cardiac muscle stretch and pressure overload, to identify patients with TACO. BNP has been shown to assist in diagnosing patients with congestive heart failure (CHF), a pathologic fluid-overload state. When comparing patients with TACO versus control patients (with other transfusion-associated reactions), this study found that a posttransfusion-to-pretransfusion BNP ratio of 1.5 was the strongest independent predicator of TACO (sensitivity 81%, specificity 89%, and accuracy 87%). The authors stated that while many cases of TACO can be diagnosed without BNP testing, this test can be a useful adjunctive marker in patients with clinically unclear presentations.
A more recent paper by the Johns Hopkins Medical Institutions Transfusion Medicine group (Tobian, et al) investigated the possibility of NT-proBNP, an inactive N-terminal complementary fragment cleaved off during the production of BNP (see Figure 1), as a useful marker in diagnosing TACO. NT-proBNP has also been found to provide diagnostic and prognostic information in patients with CHF. While both BNP and NT-proBNP are present initially in equal molar concentrations after cleavage, BNP has been found to degrade in blood specimens by more than 30% after 24 hours post-draw, resulting in possible falsely-elevated posttransfusion-to-pretransfusion BNP ratios. On the other hand, NT-proBNP is relatively stable at room temperature or 4 degrees Centigrade. Also, BNP in vivo is cleared by natriuretic peptide clearance receptors and endopeptidases, whereas NT-proBNP is cleared by renal filtration. This correlates with a much longer half-life for NT-proBNP (90 minutes) compared to BNP (20 minutes).
A case-control study was set up identifying patients with TACO versus other transfusion-reactions and measuring their NT-proBNP levels pretransfusion and posttransfusion. First, it was found that patients with TACO had significantly higher concentrations of NT-proBNP compared to control patients. This indicates that TACO patients are in an excess fluid state prior to transfusion. Even though 63% of TACO patients had an increase in their posttransfusion NT-proBNP levels compared to pretransfusion, for the aggregate group of TACO patients there was no such increase. No patients in the control group had a higher posttransfusion NT-proBNP compared to pretransfusion levels.
Second, by analyzing ROC curves (plotting sensitivity against 1-specificity) it was found that the posttransfusion NT-proBNP level was a better marker for TACO and not the posttransfusion-to-pretransfusion NT-proBNP ratio. Accuracy was optimized at a posttransfusion NT-proBNP concentration of 1923 pg/mL (sensitivity 87.5%, specificity 95.8%, and accuracy 92.5%). Additionally, multivariate logistic regression analysis demonstrated that an elevated posttransfusion NT-proBNP level greater than 1000 pg/mL was the only independent variable for the diagnosis of TACO (and not other clinical indicators).
This study demonstrated that elevated posttransfusion NT-proBNP level was sensitive, specific, and accurate for diagnosing TACO. Even though an increased posttransfusion-to-pretransfusion NT-proBNP ratio was not found to be as effective as posttransfusion levels alone, 63% of TACO patients did show posttransfusion-to-pretransfusion ratios greater than 1.0 with levels greater than 1000 pg/mL after transfusion. No patients in the control group did this. Thus, in patients that meet these criteria, NT-proBNP can support the diagnosis of TACO.
In our patient, TRALI was considerably less likely given the fact that the transfusions were packed red cells which contain only 40-50 ml of plasma and all four of the blood donors were healthy males - females, specifically females with children, are at a much higher risk of developing antibodies to HLA antigens compared to males. Also, our patient was known to be hypertensive with renal failure and thus at increased risk for developing symptomatic volume overload. He was infused with approximately 10 liters of fluid within a short time interval, increasing his risk of pulmonary edema secondary to transfusion products and other intravenous fluids. His elevated blood pressure and heart rate also strongly supports TACO.
While we have no BNP or NT-proBNP levels before or after the transfusions, this case helps to illustrate the potentially lethal outcomes of TACO. If this patient had lived, the measurement of NT-proBNP concentration may have helped support the diagnosis of TACO and appropriately direct therapeutic management. He suffered respiratory failure from pulmonary edema, congestion and intra-alveolar hemorrhage following his renal transplantation for his underlying end-stage renal disease. Respiratory failure due fluid resuscitation and transfusion are the most likely cause of death in this case.
Andrzejewski, C, et al. Transfusion-associated adverse pulmonary sequelae: widening our perspective. Transfusion 2005:45:1048-1050.
Bierbaum, B, et al. An Analysis of Blood Management in Patients Having a Total Hip or Knee Arthroplasty. The Journal of Bone and Joint Surgery 1999;81-A(1):2-10.
Fiebig, E, et al. Transfusion-related acute lung injury and transfusion-associated circulatory overload: mutually exclusive or coexisting entities? Transfusion 2007;47:171-172.
Popovsky, M, et al. Transfusion-associated circulatory overload in orthopedic surgery patients: a multi-institutional study. Immunohematology 1996;12:87-89.
Tobian, A, et al. N-terminal pro-brain natriuretic peptide is a useful diagnostic marker for transfusion-associated circulatory overload. Transfusion 2008;48:1143-1150.
Zhou, L, et al. Use of B-natriuretic peptide as a diagnostic marker in the differential diagnosis of transfusion-associated circulatory overload. Transfusion 2005;45:1056-1063.
Contributed by Jay S. Raval, MD and Darrell Triulzi, MD