Hospital Medicine

Transfusion-related acute lung injury

Jump to Section

Transfusion-related acute lung injury

I. What every physician needs to know.

Transfusion-related acute lung injury (TRALI) denotes acute respiratory distress that develops within 6 hours of a transfusion of a plasma-rich blood product. Physiologically it is similar to other forms of acute lung injury (ALI) or acute respiratory distress syndrome (ARDS) and typically presents as acute dyspnea due to noncardiogenic pulmonary edema with bilateral infiltrates on chest radiograph.

Brittingham first noted the relationship between symptoms of acute lung injury, transfusion, and leukoagglutinins in 1957. However, it was not until the early 1980's that Popovsky and colleagues described the first series of cases and the relationship to leukoagglutinins, which they termed "transfusion-related acute lung injury". Prior to 1985, other nomenclature used to describe this syndrome included noncardiogenic pulmonary edema, allergic pulmonary edema, pulmonary leukoagglutinin reaction, and pulmonary hypersensitivity reaction.

Although the etiology of TRALI has not been fully delineated, two hypotheses have been postulated. The antibody-mediated hypothesis proposes that antibodies react with a corresponding antigen triggering capillary leak. Identified antibodies include antibodies to human leukocyte antigen (HLA) class I and II antigens and human neutrophil antigen (HNA). In this hypothesis, antibodies bind to recipient neutrophils. The antibody bound neutrophils are then sequestered in the lungs where activation of complement results in endothelial damage, capillary leak, and ALI.

In most of these cases, the antibody is found in the donor with the corresponding antigen identified on the recipient's neutrophils. Most donors associated with cases of TRALI are multiparous women who become alloimmunized during pregnancy. One notable case supporting the antibody mediated hypothesis describes a patient who underwent lung transplantation and developed dyspnea and hypoxia after receiving a transfusion of two units of packed red blood cells. Chest radiograph revealed unilateral white-out in the transplanted lung alone. Antibodies to HLA-B44 were noted in the second unit of transfused blood, and the corresponding antigen was expressed on the transplanted (affected) lung but not in the native (uninvolved) lung.

In approximately 10% of TRALI cases, the antigen appears to originate in the recipient and is directed against transfused neutrophils. A small number of cases have also been reported in which both antibody and antigen are found in separate products transfused into the same individual.

Despite evidence supporting the antibody-mediated hypothesis, up to 15% of cases of TRALI have failed to identify antibodies. In addition, while HLA antibodies are commonly identified in female donors, only a very small percent of these donors are implicated in TRALI. Furthermore, transfusion of blood products from donors with known HLA antibodies to recipients with corresponding antigens causes lung injury in only some of the recipients. Also, corresponding antigens to leukocyte antibodies are not always detected in the donor. Finally, cases of TRALI after autologous transfusion have also been reported suggesting that HLA and HNA antigens are not involved in all cases of TRALI since an autologous donor unit would not contain antibodies to the donor's antigens.

These findings have led to the postulation of a second hypothesis suggesting that biologic modifiers which accumulate over time in stored blood products may also be responsible for the development of TRALI. TRALI occurring as a result of biologic modifiers seems to be associated with a more benign presentation than cases occurring as a result of antibodies.

Regardless of whether TRALI results from antibodies or from biologic modifiers, there is a growing consensus that two distinct events may be necessary for the development of TRALI. This has been termed the "two-hit" model. The first event involves an inflammatory condition (e.g., infection, trauma) which leads to sequestration and priming of neutrophils in the lungs. The second event is the transfusion. Antibodies, endotoxins, or other substances that have accumulated during blood storage stimulate the primed neutrophils to release oxidases and proteases, causing damage to the endothelium resulting in capillary leak and ALI.

This hypothesis is supported by studies in which bioactive lipids and outdated blood products have been used to cause TRALI after a priming event. One limitation of the two-hit model is that not all patients who develop TRALI were in poor health prior to receiving the blood product transfusion. A threshold model (based on the two-hit model) has been suggested by Bux and Sachs, in which a certain threshold must be surpassed in order to induce a TRALI reaction.

More recently, Middleburg and van der Bom, believing that TRALI is multifactorial in nature, have suggested reconsidering the two-hit model and have proposed two alternative models: a new type of threshold model and the sufficient cause model. In their revised threshold model, they propose that multiple causal factors combine to achieve a certain threshold resulting in TRALI. In their sufficient cause model, individual contributing factors (which they term component causes) combine in ways sufficient to cause disease. In this model, transfusion is considered a “necessary” cause as, by definition, it must be a component cause in all cases of TRALI. In October 2003 the United Kingdom began excluding multiparous women (who become alloimmunized during pregnancy) from plasma donation in order to minimize the risk of TRALI. Since then the reported cases of TRALI in the UK has decreased by almost two-thirds, and the risk of death from TRALI has almost disappeared. In the United States, after introduction of low-risk plasma in 2007, there was a 63% reduction in reported deaths. Red blood cells, despite being considered a low-risk product, now account for the largest proportion of TRALI cases in the United States.

The American Association of Blood Banks (AABB) adopted a similar standard, effective April 1, 2014. Published in the 29th edition of Standards for Blood Banks and Transfusion Services (BBTS Standards), standard 5.4.1.2 states that “plasma and whole blood for allogenic transfusion shall be from males, females who have not been pregnant, or females who have been tested since their most recent pregnancy and results interpreted as negative for HLA antibodies.” TRALI mitigation strategies as required by the AABB are limited to whole blood and plasma and do not currently apply to platelet or red blood cell products due to concerns that this strategy could result in donor shortages of platelets and red blood cell products.

Finland, Luxembourg, and Norway now use solvent/detergent (S/D)-treated fresh frozen plasma. S/D-treated plasma is plasma that is pooled from 500-1600 donors, which is then treated with solvent and detergent to inactivate lipid-enveloped viruses, cells, and most protozoa, while leaving labile coagulation factors intact. The pooling is thought to cause dilution of antibodies by at least 500 times. In addition, antigens in the plasma may also bind antibodies in the plasma rendering the antibodies inactive. Since using S/D-treated plasma, there have not been any reported cases of TRALI in these three countries.

II. Diagnostic Confirmation: Are you sure your patient has Transfusion-Related Acute Lung Injury?

TRALI is a clinical diagnosis. There are no biomarkers that facilitate diagnosis.

A case definition of TRALI based on clinical and radiological parameters was devised by the U.S. National Heart, Lung, and Blood Institute in 2004. The following criteria are necessary to make a diagnosis of TRALI or possible TRALI:

TRALI:

  • Acute lung injury (ALI)

    • Acute onset

    • Hypoxemia

      • PaO2/FiO2<300mm Hg

      • Oxygen saturation <90% on room air

      • Other evidence of hypoxia

    • Bilateral infiltrates on chest radiograph

    • No evidence of pulmonary edema (i.e., pulmonary artery wedge pressure less than or equal to 18mm Hg)

  • No preexisting ALI prior to the transfusion

  • Symptoms occurring during or within 6 hours of the transfusion

  • No temporal relationship to an alternate risk factor for ALI

Possible TRALI:

  • All of the above but another risk factor for ALI present

It has recently been suggested that TRALI and possible TRALI may in fact represent two entirely different clinical entities. A prospective observational study of hypoxemia following transfusion found that transfusion factors were not associated with possible TRALI. Specifically, neither receipt of plasma from female donors nor a larger number of units transfused were risk factors for possible TRALI. In addition, development of possible TRALI was associated with recipient risk factors (including chronic alcohol abuse, current smoking, shock before transfusion, and larger fluid balance prior to transfusion) known to increase the risk of ARDS.

Other studies have demonstrated that the known mortality of ARDS is similar to the observed mortality of possible TRALI suggesting that mortality is driven by ARDS risk factors and not by transfusion risk factors. In addition, the occurrence of possible TRALI has not been related to the HLA antibody status of the blood product donor. Finally, while the use of male-predominant plasma has decreased the rate of TRALI, it has not decreased the rate of possible TRALI. These findings support the theory that possible TRALI may, in fact, not be TRALI (i.e., due to transfused factors) but rather ARDS occurring in individuals who receive transfusions (i.e., “transfused ARDS’).

A. History Part I: Pattern Recognition:

TRALI has an acute presentation. Symptoms typically occur within 1 to 2 hours after transfusion. However, symptoms have been reported to occur during blood product infusion and can also be delayed for up to 6 hours post-transfusion. It has also recently been recognized that in critically ill or injured patients, ALI can develop as late as 6-72 hours following the administration of blood products. This ‘delayed’ form of TRALI is associated with a high mortality rate.

Most patients present with acute respiratory distress, but the presentation of TRALI can range from transient dyspnea to fulminant acute lung injury/acute respiratory distress. Patients most commonly present with dyspnea, cough, and fever. Other signs and symptoms include cyanosis, hypertension, and hypotension. Patients who are intubated may also develop large amounts of frothy fluid from the endotracheal tube.

Chest radiographs typically demonstrate diffuse bilateral infiltrates, often described as a "white-out" pattern but in the first few hours can show a patchy pattern.

B. History Part 2: Prevalence:

Although TRALI remains rare, it has been the leading cause of transfusion-related death reported to the FDA since 2001. This increase is likely due to prior underdiagnosis and under-reporting of cases of TRALI and not due to an increase in the number of cases of TRALI. The fatality rate of 5-20% associated with TRALI is the highest for all transfusion reactions, and in the intensive care setting, the mortality rate may be as high as 47%. The incidence of TRALI varies widely due to varying definitions used. TRALI is the most common adverse reaction associated with transfusion of blood products with reported incidences in North America ranging from 1 in 5,000 to 1 in 1,323 transfusions and reported incidences in Europe ranging from 1.3 in 1,000,000 to 1 in 7,900.

Of the 176 transfusion-related fatalities reported to the US Food and Drug Administration (FDA) between fiscal years 2010 and 2014, 72 (41%) were due to TRALI. TRALI occurs with similar frequencies in males and females and has been reported in patients of all ages. All blood products containing plasma including fresh frozen plasma, platelets (both apheresis and concentrates), packed red blood cells, whole blood, cryoprecipitate, and granulocytes, have been associated with cases of TRALI. There have also been three case reports of TRALI after infusion of intravenous immune globulin. TRALI has been reported in patients who received as little as 10-15 ml of plasma, and surveillance data studies suggest that products rich in plasma, such as fresh frozen plasma and apheresis platelets, have the highest associated risk of TRALI.

C. History Part 3: Competing diagnoses that can mimic Transfusion-Related Acute Lung Injury.

It can often be difficult to distinguish between TRALI and transfusion-associated circulatory overload (TACO) as both entities present with respiratory distress and bilateral pulmonary edema. However, unlike pulmonary edema, with TRALI the pulmonary wedge pressure should not exceed 18mmg Hg.

Processes that can present similarly to TRALI include cardiogenic pulmonary edema (including TACO and myocardial infarction), acute hemolytic transfusion reactions, anaphylactic transfusion reactions, and bacterial sepsis from a contaminated blood component. TACO can develop within minutes to hours post-transfusion and usually presents with tachypnea, tachycardia, cyanosis, and hypertension. However, with TACO and with cardiogenic pulmonary edema, jugular venous pressure and pulmonary capillary wedge pressure are elevated. An S3 gallop may also be present.

Anaphylactic transfusion reactions typically present with wheezing due to bronchospasm, severe hypotension, and erythema and urticaria involving the head, neck, and trunk. Bacterial sepsis usually presents with fever which can be seen with TRALI; respiratory distress is uncommon in bacterial sepsis but can present as a manifestation of vascular collapse. Hemolytic transfusion reactions may present with symptoms somewhat similar to TRALI but can be easily distinguished by the presence of hemolysis.

D. Physical Examination Findings.

Auscultation of the lungs reveals diffuse crackles and decreased breath sounds, especially in dependent areas. Patients with TRALI do not have elevated jugular venous distension or an S3 gallop.

E. What diagnostic tests should be performed?

Measurement of pulmonary capillary wedge pressure can be helpful in distinguishing between TRALI and TACO; however, pulmonary artery catheterization should be performed only if clinically indicated.

1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

A complete blood count (CBC) obtained in the acute setting can be helpful as transient leukopenia has been reported.

Demonstration of antibodies against HLA class I or II or HNA antigens in the donor plasma with identification of the corresponding antigen on recipient neutrophils is useful in confirming the diagnosis of TRALI but is not necessary for the diagnosis.

2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

Chest X-ray typically reveals bilateral alveolar and interstitial infiltrates, consistent with bilateral pulmonary edema.

F. Over-utilized or “wasted” diagnostic tests associated with this diagnosis.

More recent studies have suggested that brain natriuretic peptide (BNP) is not useful when attempting to distinguish between TRALI and TACO because hypoxemia in the setting of TRALI can also stimulate secretion of BNP.

III. Default Management.

N/A

A. Immediate management.

When TRALI is suspected, the transfusion should be immediately discontinued. Management is largely supportive. Most patients will require supplemental oxygenation, and approximately 70% of patients will require mechanical ventilation.

In ventilated patients pulmonary fluid should be collected from the endotracheal tube when possible, early in the course of lung injury. Simultaneously measuring pulmonary fluid and plasma protein can assist in the diagnosis of increased permeability pulmonary edema. For example, with hydrostatic pulmonary edema, the edema fluid/plasma protein ratio is <0.65, while with increased permeability pulmonary edema the ratio is >0.75. Of note, this technique is valid only from undiluted pulmonary fluid and should not be performed via bronchoalveolar lavage.

A CBC should be obtained as transient leukopenia, which may be the result of massive sequestration of leukocytes in the pulmonary vasculature, can be seen with TRALI.

Because the pulmonary edema in TRALI is not due to volume overload, diuretics (which can result in decreased cardiac output) should be avoided. In cases where the diagnosis remains uncertain, echocardiogram and/or pulmonary artery catheterization may be helpful in distinguishing between TRALI and TACO.

The blood bank should be notified. The recently transfused blood bag should be retained and sent to the blood bank along with blood drawn from the patient. The blood bank should investigate the transfused blood products within 4-6 hours of the transfusion. Cases of suspected TRALI should be reported to the National Blood Bank for serologic testing of the recipient and implicated donors. Donors implicated in possible cases of TRALI should be temporarily disqualified from blood donation pending testing for leukocyte antigens. A donor who is noted to have antibodies that are incompatible with the recipient is then disqualified from future plasma or platelet donation.

B. Physical Examination Tips to Guide Management.

Improving rales suggests clinical improvement of TRALI.

C. Laboratory Tests to Monitor Response To, and Adjustments in, Management.

NA

D. Long-term management.

Management is largely supportive, and most patients will require mechanical ventilation.

E. Common Pitfalls and Side-Effects of Management.

In patients who develop hypoxemia with bilateral pulmonary infiltrates in the setting of transfusion of blood products, TRALI must be differentiated from TACO. Other causes of ALI need also to be considered.

IV. Management with Co-Morbidities.

N/A

A. Renal Insufficiency.

No change in standard management.

B. Liver Insufficiency.

No change in standard management.

C. Systolic and Diastolic Heart Failure.

TRALI must be differentiated from TACO as outlined previously.

D. Coronary Artery Disease or Peripheral Vascular Disease.

TRALI must be differentiated from myocardial infarction as myocardial infarction with pulmonary edema can present similarly.

E. Diabetes or other Endocrine issues.

No change in standard management.

F. Malignancy.

No change in standard management.

G. Immunosuppression (HIV, chronic steroids, etc).

No change in standard management.

H. Primary Lung Disease (COPD, Asthma, ILD).

A diagnosis of TRALI cannot be made in patients who have pre-existing ALI.

I. Gastrointestinal or Nutrition Issues.

No change in standard management.

J. Hematologic or Coagulation Issues.

No change in standard management.

K. Dementia or Psychiatric Illness/Treatment.

No change in standard management.

V. Transitions of Care.

A. Sign-out considerations While Hospitalized.

The hospitalist should have a low threshold to transfer a patient with suspected TRALI to the intensive care unit as 70% of patients with TRALI will require mechanical ventilation.

B. Anticipated Length of Stay.

With appropriate respiratory support, 80% of patients with TRALI show dramatic improvement within 48 to 96 hours after the insult. Chest radiographs demonstrate improvement in pulmonary edema in a similar time frame. The remaining 20% of patients have a more protracted course with pulmonary infiltrates persisting for more than 7 days.

In contrast to patients with ALI who have a mortality rate of 40-60%, the mortality rate in patients with TRALI is approximately 5-10%. Early fatalities occur due to the inability to maintain adequate ventilation in the setting of significant pulmonary edema. Late fatalities are typically due to complications of ventilator therapy such as pneumonia.

C. When is the Patient Ready for Discharge.

The patient is ready for discharge when he is no longer hypoxic and the underlying medical issue for which the patient received the transfusion does not require the patient to remain hospitalized.

D. Arranging for Clinic Follow-up.

N/A

1. When should clinic follow up be arranged and with whom.

Routine follow-up with a hematologist is not necessary.

2. What tests should be conducted prior to discharge to enable best clinic first visit.

None

3. What tests should be ordered as an outpatient prior to, or on the day of, the clinic visit.

None

E. Placement Considerations.

NA

F. Prognosis and Patient Counseling.

There are no known long term sequelae of TRALI. Patients who recover usually regain full pulmonary function, without apparent structural damage. This is in contrast to ALI in which many patients develop irreversible lung injury.

However, as recurrent cases of TRALI have been reported, further transfusions in patients with TRALI reactions should be performed only after careful consideration.

VI. Patient Safety and Quality Measures.

A. Core Indicator Standards and Documentation.

NA

B. Appropriate Prophylaxis and Other Measures to Prevent Readmission.

No change in standard management.

VII. What's the evidence?

"TRALI Risk Mitigation for Plasma and Whole Blood for Allogenic Transfusion. Bulletin #14-02". January 29, 2014. http://www.aabb.org/programs/publications/bulletins/Documents/ab14-02.pdf.

Droog, W, van Thiel, SW, Sleeswijk Visser, SJ, Rhemrev, JPT.. "Complications Due to Transfusion-Related Acute Lung Injury". Obstet Gynecol.. vol. 113. 2009. pp. 560-563.

El Kenz, H, Van der Linden, P.. "Transfusion-Related Acute Lung Injury". Eur J Anaesthesiol.. vol. 31. 2014. pp. 345-350.

Kopko, PM.. "Transfusion-Related Acute Lung Injury". J Infus Nurs.. vol. 33. 2010. pp. 32-37.

Lee, JH, Kang, ES, Kim, DW.. "Two Cases of Transfusion-Related Acute Lung Injury Triggered by HLA and Anti-HLA Antibody Reaction". j Korean Med Sci.. vol. 25. 2010. pp. 1398-1403.

Looney, MR, Gropper, MA, Matthay, MA.. "Transfusion-Related Acute Lung Injury A Review". Chest.. vol. 126. 2004. pp. 249-258.

Middleburg, RA, van der Bom, JG.. "Transfusion-Related Acute Lung Injury Not a Two-Hit But a Multicausal Model". vol. 55. 2015. pp. 953-960.

Peters, AL, Van Stein, D, Vlaar, APJ.. "Antibody-Mediated Transfusion-Related Acute Lung Injury; From Discovery to Prevention". British Journal of Haematology.. vol. 170. 2015. pp. 597-614.

Popovsky, M.. "Tranfusion-Related Acute Lung Injury: Three decades of Progress But Miles To Go Before We Sleep". Transfusion.. vol. 55. 2015. pp. 930-934.

Quest, GR, Gaal, H, Clarke, G, Nahirniak, S.. "Tranfusion-Related Acute Lung Injury after Transfusion of Pooled Immune Globulin: A Case Report". Transfusion.. vol. 54. 2014. pp. 3088-3091.

Silliman, CC, Ambruso, DR, Boshkov, LK.. "Transfusion-Related Acute Lung Injury". Blood.. vol. 105. 2005. pp. 2266-2273.

Toy, P, Bacchetti, P, Grimes, B. "Recipient Clinical Risk Factors Predominate in Possible Transfusion-Related Acute Lung Injury". Transfusion. vol. 55. 2015. pp. 947-952.

Triulzi, DJ.. "Transfusion-Related Acute Lung Injury: Current Concepts for the Clinician". Anesth Analg.. vol. 108. 2009. pp. 770-776.

Triulzi, DJ.. "Transfusion-Related Acute Lung Injury: An Update". Hematology. 2006. pp. 497-501.

"Fatalities Reported to FDA Following Blood Collection and Transfusion. Annual Summary for Fiscal Year 2014". Silver Spring, MD. August 2015. http://www.fda.gov/downloads/BiologicsBloodVaccines/SafetyAvailability/ReportaProblem/TransfusionDonationFatalities/UCM459461.pdf.

Vlaar, AP, Schulz, MJ, Juffermans, NP.. "Transfusion-Related Acute Lung Injury: A Change of Perspective". Neth J Med.. vol. 67. 2009. pp. 320-326.

You must be a registered member of Clinical Advisor to post a comment.
close

Next Article in Hospital Medicine

Sign Up for Free e-newsletters