Article Text
PDF
Abstract
We present a never-transfused girl with thalassemia intermedia who was admitted for febrile aplastic crisis due to human parvovirus B19. After a first transfusion of packed red blood cells, she developed pulmonary oedema. She improved with supportive care including the use of intravenous diuretics. Due to severe anaemia, she received a second blood transfusion, antibiotics for febrile neutropenia and intravenous γ globulin for control of the parvovirus infection. She had an uneventful recovery. The first of her male blood donors had an antibody against a patient’s human leukocyte antigens type II antigen with a high mean fluorescent intensity. Our patient had clinical features and supportive laboratory evidence for mild transfusion-related acute lung injury (TRALI). However, she also met the criteria for transfusion-associated circulatory overload (TACO). We conclude that our patient likely suffered from TRALI/TACO, a consensus term proposed in 2019 for patients in whom TRALI cannot be distinguished from TACO or in whom both conditions occur simultaneously.
- Haematology (drugs and medicines)
- Paediatrics (drugs and medicines)
Statistics from Altmetric.com
Background
Respiratory transfusion reactions include transfusion-related acute lung injury (TRALI) and transfusion-associated circulatory overload (TACO) (Box 1).1 TRALI is a form of acute respiratory distress syndrome (ARDS) occurring during or within 6 hours of transfusion of any blood product that contains plasma (plasma, packed red blood cells (RBCs), platelets). There is compelling accumulating evidence that transfusing plasma-rich blood components from male donors or donors who tested negative for leukocyte antibodies effectively reduces TRALI.2
New consensus TRALI definition and definition of TACO, as per the National Healthcare safety network. Note the proposed TRALI/TACO category.
Diagnostic criteria for TRALI
TRALI type I: No known risk factors for acute respiratory distress syndrome (ARDS).
Diagnostic characteristics
Acute onset.
Hypoxemia.
Bilateral pulmonary oedema on imaging.
No left atrial hypertension or left atrial hypertension present and judged not to be the main contributor to the hypoxemia.
Timing of TRALI symptoms relative to transfusion
Onset during or within 6 hours of transfusion.
Patient status before symptom onset
No temporal relationship to an alternative risk factor for ARDS.
TRALI type II: Risk factors for ARDS are present.
Diagnostic characteristics
As shown above.
Timing of TRALI symptoms relative to transfusion
Onset during or within 6 hours of transfusion.
Patient status before symptom onset
Stable respiratory function in the 12 hours before transfusion.
Diagnostic criteria for TACO as per the National Healthcare Safety Network*
Definitive TACO: New onset or exacerbation of three or more of the following within 12 hours of cessation of transfusion (at least one of the following):
Evidence of acute or worsening respiratory distress (dyspnoea, tachypnoea, cyanosis and decreased oxygen saturation values in the absence of other specific causes); and/or
Radiographical or clinical evidence of acute or worsening pulmonary oedema (crackles on lung auscultation, orthopnoea, cough, a third heart sound and pinkish frothy sputum in severe cases); or both AND
Elevated brain natriuretic peptide or NT-proBNP relevant biomarker;
Evidence of cardiovascular system changes not explained by underlying medical condition (elevated central venous pressure, evidence of left heart failure including development of tachycardia, hypertension, widened pulse pressure, jugular venous distension, enlarged cardiac silhouette and/or peripheral oedema);
Evidence of fluid overload.
Probable TACO: N/A
Possible TACO: N/A
*Available online at: https://www.cdc.gov/nhsn/pdfs/biovigilance/bv-hv-protocol-current.pdf
Classification of pulmonary oedema not fulfilling TRALI criteria
ARDS: Patients who have risk factors for ARDS and deteriorate not due to transfusion, but as a consequence of the already present ARDS risk factors.
Onset of ARDS within 6 hours after transfusion but respiratory status was deteriorating in the 12 hours before transfusion.
Existing ARDS of any severity that further deteriorates after transfusion where respiratory status was already deteriorating in the 12 hours before transfusion.
TRALI/TACO cannot be distinguished: Patients in whom TRALI cannot be distinguished from TACO or in whom both conditions occur simultaneously. Clinical findings compatible with TRALI and with TACO and/or lack of data to establish whether significant left atrial hypertension is present.
ARDS, acute respiratory distress syndrome; NT-proBNP, N-terminal prohormone of brain natriuretic peptide; TACO, transfusion-associated circulatory overload; TRALI, transfusion-related acute lung injury.
The consensus definition of TRALI was updated in 2019.3 It distinguishes type I (without risk factor for ARDS) from type II (with risk factor for ARDS) TRALI to better consider the possibility of the coexistence of TRALI with other risk factors for respiratory distress. In addition, the term ‘possible TRALI’ has been removed, and a new category namely TRALI/TACO has been included for those patients in whom TRALI cannot be distinguished from TACO or in whom both conditions occur simultaneously.3 In TRALI, the fluid of pulmonary oedema is typically protein rich. Diagnosis of TRALI is based on clinical and radiological signs alone. Since TRALI can result from donor-derived alloantibodies that attack class I or II human leukocyte antigens (HLA) or human neutrophil antigens (HNA) on the recipient’s blood cells, demonstration of such alloantibodies supports the diagnosis of TRALI. However, in approximately 20% of TRALI cases, HNA and HLA antibodies cannot be identified in the transfused product. These cases of ‘non-immune’ TRALI may be due to exposure to other biological reactive molecules also known as biological response modifiers (BRMs) within the transfused blood products.4 BRMs are thought to be the necessary triggers of the cascade of pulmonary injury in a primed transfused recipient. TRALI remains a clinical diagnosis, and detection of cognate anti-HLA or anti-HNA antibodies is not required, especially since the presence of incompatible alloantibodies does not always provoke TRALI.5 6
On the other hand, the hallmark of TACO is the acute generation of protein-poor pulmonary oedema in association with volume overload from rapid transfusion of blood volume that exceeds the patient’s cardiovascular capacity to handle it, typically but not exclusively, in the context of cardiac and/or renal failure. TACO is an under-reported diagnosis, and clinical descriptions include evidence of positive fluid balance or cardiac involvement with left heart failure. Circulatory overload in TACO is manifested by weight gain, oedema, increased jugular vein pressures, increased natriuretic peptides, a higher probability of hypertension and protein-poor and cell-free pulmonary oedema that responds to treatment with diuretics.7 In 2018, the definition of TACO in the international surveillance case was revised.8 For the diagnosis of TACO, evidence of acute or worsening respiratory distress, acute or worsening pulmonary oedema, or both are required. Additional criteria, that is, evidence for cardiovascular system changes, evidence of fluid overload and elevation in B-type natriuretic peptide concentrations, or a combination of these are necessary. A total of ≥3 criteria is required for diagnosis. This revised TACO surveillance case definition captures 76% of cases endorsed as TACO by participating hemovigilance systems.8 The detailed definition of TACO, as per the most recent document of the National Healthcare Safety Network (February 2023), is shown in Box 1.
Case presentation
A well-known, never-transfused girl in middle childhood with thalassemia intermedia, was admitted to our clinic with a fever of up to 39.7°C for 5 days and decreased oral intake. The patient’s haemoglobin genotype includes compound heterozygosity of haemoglobin O Arab, along with the CD5 mutation of β-thalassemia and coinheritance of a triplicated anti-α3.7 gene.
On admission, she was pale, 40 breaths per minute, 142 beats per minute with a 2/6 systolic ejection heart murmur and palpable spleen 5 cm below the left costal margin. The spleen size was unchanged from the patient’s baseline physical examination. A haemogram on admission showed haemoglobin 40 g/L, down from a baseline value of 76 g/L, leukocytes 1.76 x 109/L, absolute neutrophil count 0.211 x 109/L, platelets 74 x 109/L and reticulocytes 0.24%. A chest radiograph on admission showed a subtle pleural effusion on the left haemithorax (figure 1a). Biochemical testing revealed total bilirubin 0.9 mg/dL, lactate dehydrogenase 527 U/L, normal transaminases, amylase, γGT, urea and creatinine, and C-reactive protein 0.8 mg/dL. Serological testing detected IgM antibodies for parvovirus B19, consistent with an acute infection.
Due to febrile neutropenia, the patient was started on empiric intravenous piperacillin/tazobactam pending blood culture results that were eventually negative. Because of the severe anaemia, she was transfused with 11 mL/kg of packed RBCs over 2 hours from a male donor. Towards the end of the transfusion, the doctors on call noted distension of the right jugular vein and worsening tachypnoea along with hypoxemia needing 2 litres of supplemental oxygen by face mask to maintain oxygen saturation >94%. Bilateral rales and wheezing were also observed, for which she was prescribed systematic intravenous furosemide 0.5 mg/kg every 12 hours for possible TACO along with nebulized salbutamol. Her blood pressure was normal (100/60 mm Hg), and she did not develop peripheral oedema. Approximately 38 hours after the first transfusion, due to inadequate haemoglobin elevation (post-transfusion haemoglobin 51 g/L), the patient, who remained febrile and pancytopenic, received a second transfusion with 12 mL/kg of packed RBCs over 4 hours from a different male donor. The response was good, with post-transfusion haemoglobin 72 g/L. At the same, she continued to receive intravenous furosemide.
Investigations
An urgent echocardiogram after the first transfusion showed good cardiac contractility. A new portable chest radiograph obtained on a semi-sitting position 4.5 hours after the second transfusion demonstrated bilateral pulmonary oedema with pleural effusions, a significant deterioration compared with the initial radiograph (figure 1b). Due to the need for continuous administration of supplemental oxygen despite the use of systemic diuretics, TACO/TRALI and TRALI were strongly considered at this point, and blood samples from the patient and the two donors were sent to the National Histocompatibility Center for HLA typing of the patient and to be screened by Luminex for HNA and HLA antibodies against the patient’s blood cell antigens.9 In addition, N-terminal prohormone of brain natriuretic peptide (NT-proBNP) was measured in the patient’s serum and was found to be 1202 pg/mL, a value approximately five times higher than the upper limit of the published reference range for her age.10 The Luminex HNA screening (LSMulti, One-Lambda) was negative, while the single-antigen bead assay detected antibodies against the patient’s HLA-DQB1*05:01, with a mean fluorescent intensity (MFI) of 2493 (MFI cut-off 1000) (figure 2). The time course of all medical events since admission is summarised in table 1.
(a) Initial chest radiograph on admission. The hardly noticeable pleural effusion on the left (blue arrow). (b) The second chest radiograph. It was obtained 4.5 hours after the second transfusion in a semi-sitting position with a portable X-ray machine. Note the bilateral pleural effusions and pulmonary oedema.
Testing for human leukocyte antigens (HLA) class II antibodies using Luminex Single Antigen Beads assay (one lambda). Antibodies in the plasma of the first donor against HLA-DQB1*05:01 of the patient were detected with a mean fluorescent intensity of 2493.
Time course of all medical events since admission
Differential diagnosis
Pneumonia due to human parvovirus B19 and/or other viral pathogens was initially considered. The subtle left pleural effusion on admission was likely due to the underlying parvovirus B19 infection in the context of severe hyporegenerative anaemia with critically low haemoglobin. Given the negative blood culture results and the low serum CRP, bacterial pneumonia was deemed highly unlikely. Despite that, antibiotics were prescribed due to the parvovirus B19-related febrile pancytopenia (aplastic crisis), a well-known complication of this virus in patients with haemolytic anaemias.11 TACO (cardiogenic pulmonary oedema) and TRALI (non-cardiogenic pulmonary oedema) were distinct possibilities in our never-transfused patient, who developed post-transfusion pulmonary oedema. Our initial diagnostic thought was that our patient suffered from TACO, as patients with severe anaemia may be prone to volume overload, especially in the setting of a rapid transfusion. This is why the second chest radiograph was delayed for approximately 2 days after the initial respiratory compromise hoping that administration of systemic diuretics would abate further respiratory deterioration.
The good cardiac contractility, as documented by cardiac ultrasound, the normal blood pressure, the absence of peripheral oedema and the positive HLA screen for antibodies in the blood of the first donor against the patient’s HLA-DQB1*05:01 are suggestive of TRALI. However, they cannot exclude TACO since not all diagnostic criteria for TACO should be present in every case. In favour of TACO is the good response to diuretics, the distension of the right jugular vein after the initial transfusion and the fact that the patient’s leukocytes did not decline further after the first transfusion—something typically seen in patients with TRALI. However, we should note that the patient was already leukopenic due to the acute parvovirus B19 infection. Other distinct causes of ARDS, such as acute pancreatitis, aspiration and multiple trauma, were ruled out by history, physical examination and incompatible laboratory evidence.
Treatment
Due to persistent febrile pancytopenia on the fifth hospital day, the patient was given a single dose of intravenous γ globulin 0.8 g/kg for control of the parvovirus B19 infection,12 with subsequent substantial clinical and haematological improvement. She defervesced and did not require supplemental oxygen anymore, and the systemic diuretics were discontinued, while the neutropenia and thrombocytopenia resolved, the reticulocyte count climbed to 5.2%, and on the seventh hospital day, she was sent home.
Outcome and follow-up
2 weeks after the patient’s discharge, a repeat NT-proBNP measurement was normal (65.4 pg/mL), as was a repeat chest radiograph. By that time, the patient’s haemoglobin had returned to its baseline value.
Discussion
We believe that our patient suffered from TRALI/TACO, as per the consensus redefinition of 2019,2 in the context of an underlying acute parvovirus B19 infection, as a possible risk factor for acute lung injury given the abnormal initial chest radiograph demonstrating a subtle left pleural effusion. Parvovirus B19 has been implicated as a causative agent of lung infections, mostly but not exclusively in immunocompromised individuals.13 The distended right jugular vein after the first transfusion and the elevated NT-proBNP that was measured after the second transfusion is more supportive of cardiogenic oedema, that is, TACO. Our patient fulfilled diagnostic criteria for TACO with evidence of worsening respiratory distress, radiographic and clinical evidence of worsening pulmonary oedema, elevated NT-proBNP and jugular venous distension. However, the need for supplemental oxygen for several days despite the use of diuretics was in favour of TRALI since slow recovery from a transfusion reaction is more typical of it. Moreover, the detection of antibodies in the blood of the first donor against an HLA-type II antigen of our patient is further supportive of the diagnosis of immune TRALI. Taking all the above into consideration, we conclude that our patient met the proposed definition of TACO/TRALI, as shown in Box 1.
Normal non-alloimmunized healthy male donors can have ‘natural’ HLA antibodies to infrequent HLA specificities, likely due to cross-reactive epitopes found in microorganisms, dietary proteins and allergens.14 The fact that our patient had a normal echocardiogram after the first transfusion and did not develop hypertension, peripheral oedema or weight gain supports TRALI. However, the systemic diuretics prescribed for the pulmonary oedema could have prevented the development of additional signs of circulatory overload. Regarding the high post-transfusion value of NT-proBNP, although its measurement is encouraged as a supportive criterion for TACO in case it is elevated >1.5 times compared with the pretransfusion value, this biomarker’s accuracy in distinguishing cardiogenic from non-cardiogenic oedema has been questioned15; plus, we did not measure it before transfusion.
The typical patient with TRALI is a critically ill ICU patient on mechanical ventilation, with a high score on the American Society of Anesthesiologists Physical Status Classification System, who receives plasma-containing blood products.16 Although our patient did not have the clinical characteristics of the typical patient with TRALI, cases of TRALI have rarely been described in transfused patients with thalassemia17 and even in normal volunteers who received anti-HLA type II antibodies.18
TRALI is thought to be a ‘two hit’ pathological process, requiring underlying inflammation with neutrophilic or vascular endothelial activation, and eventual sequestration of neutrophils in the lungs. Transfusion provides the second hit that triggers the neutrophil attack. Transfusion-derived ‘second hits’ include HNA antibodies or BRMs that accumulate on stored blood products, such as cytokines, CD40 ligand or bioactive lipids that activate neutrophils. In addition, HLA class I antibodies that bind to neutrophils, endothelial cells and, finally, HLA class II antibodies act indirectly by stimulating class II antigen–bearing cells such as macrophages. TRALI, therefore, has multiple possible aetiologies. Restricted use of blood products when only clinically indicated, deferral of blood donors implicated in TRALI reactions, screening of all donors for anti-neutrophil or anti-HLA antibodies, and use of fresher blood products may prevent TRALI. To this effect, the use of primarily male plasma donors for transfusions has been associated with a significant decrease in TRALI cases worldwide.19 20
The TRALI threshold model suggests that a certain threshold must be exceeded to induce lung injury. In mild TRALI, in which oxygen administration is sufficient, as in our case, the threshold is lower than that of severe TRALI, where patients require mechanical ventilation. Overcoming these thresholds depends on the cell-activating competence of the antileukocyte antibodies and the patient’s predisposition. A transfused strong antibody can induce a TRALI reaction in a healthy individual even if the individual predisposition is low. On the other hand, in a critically ill patient with comorbidities such as sepsis and preactivated pulmonary endothelium, the transfusion of antileukocyte antibodies with a small neutrophil–priming activity may be sufficient to overcome TRALI’s threshold.21
In conclusion, the differential diagnosis of TACO versus TRALI can be extremely difficult, as demonstrated in our female patient with thalassemia intermedia, who developed post-transfusion dyspnoea with pulmonary oedema after transfusion of packed RBCs for acute parvovirus B19 aplastic crisis. She had clinical and laboratory evidence that supported both diagnoses. We believe the 2019 consensus redefinition of TRALI is valuable, as it includes a separate category of TRALI/TACO in patients where TRALI cannot be distinguished from TACO or when both conditions can occur simultaneously as seen in our patient.
Patient’s perspective
Patient’s mother: I was told that my daughter developed a serious complication because of her first transfusion. She recovered quickly with supportive care. I wish we had come earlier to the hospital. I am indebted to the doctors for taking good care of her, and I hope this will never happen again in case she needs another transfusion in the future.
Learning points
The consensus redefinition of TRALI is useful because it includes a separate category of TRALI/TACO, in patients in whom TRALI cannot be distinguished from TACO or in whom both conditions can co-occur, like in our patient.
Although rare, TRALI/TACO can occur in never previously transfused children with thalassemia.
A high index of suspicion is required for prompt diagnosis, which is a prerequisite for a successful clinical outcome.
Ethics statements
Patient consent for publication
References
Footnotes
Contributors The following authors were responsible for drafting of the text, sourcing and editing of clinical images, investigation results, drawing original diagrams and algorithms and critical revision for important intellectual content: EM, AV, AS and EK. The following author gave final approval of the manuscript: EM.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.