Case and images courtesy of Dr. Shannon McCounnaughey.
- Initial management of a patient with suspected transfusion reaction
A 70 yo M with a history of atrial fibrillation, polycythemia vera who presents with dyspnea following a platelet transfusion. He received 1 pack of platelets in for a platelet count of 29k in anticipation of a dental procedure. Near the end of the transfusion (~ 1.5 – 2 hours after initiation of transfusion), he developed acute onset shortness of breath, shaking chills, chest tightness and palpitations. He otherwise denied any rash, pruritis, tongue or lip swelling, chest pain. Prior to the procedure, he was in his usual state of health.
His PMH is additionally notable for reactive airway disease, mildly depressed LVEF of 40-45% on most recent TTE from ~ 6 months prior. His medications include Augmentin (started for a dental infection), Advair, metoprolol, and ruxolitinib.
What would you do now?
- Stop the transfusion immediately! This most likely represents a transfusion reaction given proximity of his symptoms to his transfusion and normal state of health prior to the transfusion1.
- Confirm the patient and product – though rare, human error can lead to the wrong patient recieving the wrong product which can result in life threatening ABO blood type incompatibility and hemolysis1
- Evaluate the patient – symptoms and vital signs can help narrow the differential
- Keep IV line open with normal isotonic saline1
- Contact transfusion services/ the blood bank – most blood banks will have a protocol in place for suspected transfusion reactions which can involve sending any remaining blood product and IV tubing to the blood bank, evaluating for hemolysis,
antibody testing, etc.1
On exam, he is febrile to 38.3C, HR 164, RR 2, BP 96/30 (baseline is normal). SaO2 84% on RA, which improved to 96% on 6L NC. He has increased work of breathing, diffuse crackles in his bilateral bases, no wheezes, or rhonchi. He had no notable lip or tongue swelling and had an open posterior oropharynx. Cardiac exam is notable for JVP of 7 mmHg, no LE edema. He had scattered petechiae in his extremities but no raised wheals.
What is your differential? (~ 5 min)
This is most likely a transfusion reaction. Timing of a transfusion and different blood products can help differentiate between different types of transfusion reactions.
Based on the timing of his symptoms, this is most likely:
- Febrile non-hemolytic transfusion reaction
- Transfusion related circulatory overload (TACO)
- Transfusion related acute lung injury (TRALI)
- Transfusion related sepsis
Non-transfusion related differential includes sepsis, anaphylaxis, flash pulmonary edema from symptomatic tachyarrhythmia, reactive airway disease exacerbation.
What is your next step in evaluation and management? (~ 5 min)
Delaney, et al (2016) in Lancet, broke down the initial evaluation/management of transfusion reactions into 4 broad symptomologies – increased temperature, allergic symptoms, respiratory symptoms and other (which includes chills/rigors, nausea/vomiting and hypotension) (click for additional information)
Our patient had both a fever and respiratory symptoms. Initial work-up could include:
- Basic labs – BMP and CBC (a drop in hematocrit or platelet count can indicate a hemolytic reaction)
- Lactate given possible concern for sepsis
- Blood cultures from the patient and from platelet transfusion bag
- CXR for new oxygen requirement
- EKG for new tachycardia and oxygen requirement
- Troponin and BNP are also reasonable
Initial management could include:
- IVF initially given relative hypotension
- Empiric treatment of possible allergic reaction – epinephrine, antihistamines, and steroids
- Empiric antibiotics for possible sepsis
He received 1L IVF, IV benadryl with improvement in his BPs into normal range. He additionally received IM epinephrine and 100 mg of IV hydrocortisone for possible anaphylaxis.
Initial labs showed :
- BMP with Cr of 1.23 (baseline is normal), BUN 24.
- Lactate was 4
- CBC with WBC of 10k, hct of 26% (near baseline), platelets of 22k (platelet count was 26k prior to transfusion)
- BNP 285 (no prior baseline)
- Troponin negative
- No evidence of acute hemolytic reaction
- Gram stain on platelets transfusion bag was negative
EKG showed sinus tachycardia with HR of ~ 130 bpm
CXR showed: (click image for results)
Click here for CXR interpretation
How does this change your differential?
This CXR with bilateral infiltrates is most concerning for TRALI or TACO. Less likely flash pulmonary edema from a tachyarrhythmia (was in sinus rhythm on EKG) or new MI (troponin was negative) or hypertensive emergency (hypotensive on presentation). Transfusion related sepsis is less likely given negative gram stain.
Given the presence of right heart enlargement on CXR, a CT PE was obtained to rule out PE. There was no evidence of PE, but confirmed diffuse pulmonary edema most consistent with TRALI and less likely, TACO.
TTE ultimately showed normal LVEF of 50%, mild RV dilation and dysfunction, mild pulmonary hypertension of PA-SP 45 mmHg. RA-P was 5-10 mmHg.
Given relatively normal CVPs (JVP nonelevated on exam, RA-P within normal limits on TTE), modestly elevated BNP, and hypotension on presentation, this patient was thought to be TRALI over TACO.
How do you differentiate between these two types of pulmonary transfusion reactions?
TACO and TRALI present similarly with acute onset dsypnea, hypoxia most commonly within 6 hours of transfusion. Many risk factors are similar – including recent shock, surgery and positive fluid balance. FFP and platelets generally confer the highest risk of these reactions. TACO has higher incidence and higher mortality.
Roubinian, et al (2017) did a case-control study of pulmonary transfusion reactions (including TACO and TRALI) and found that age, BNP > 1000 pg/mL and positive fluid balance were the main risk factors that distinguished TACO from TRALI.
Physical exam findings of volume overload such as elevated JVP and LE edema are more suggestive of TACO (though patients with TRALI can develop concurrent volume overload). In addition to BNP testing, a TTE can help evaluate for elevated CVP.
Laboratory testing includes donor leukocyte antibodies, which are seen in TRALI but not typically present in TACO.
What is the management of this patient’s diagnosis?
STOP THE TRANFUSION!
Otherwise, treatment is supportive:
- Support oxygenation – if mechanical ventilation is needed, use lung protective ventilation
- Vasopressors or fluids may be needed for hypotension
- While there’s no clear role for diuresis, like in ARDS, a fluid restrictive strategy is preferred
- Some case reports have suggested clinical benefit with steroids, this has not been studied in randomized control trials3
See the Additional Learning Section below to learn more about the two hit hypothesis (pathogenesis of TRALI) and prevention measures.
TAKE HOME POINTS:
- In patients with suspected transfusion reactions, the first step is to STOP THE TRANSFUSION! Other initial steps include maintaining an IV line, confirming the patient and product and contacting the blood bank.
- Two pulmonary transfusion reactions, TRALI (akin to ARDS) and TACO (akin to CHF), present with acute onset dyspnea, hypoxia and bilateral infiltrates on CXR. They both occur most commonly within 6 hours of transfusion.
- TRALI and TACO can be differentiated by volume assessment and presence of fever. Patients with TACO have clinical signs of volume overload and a very elevated BNP. Patients with TRALI tend to be euvolemic and commonly present with fever.
- TRALI is managed supportively.
What is the two hit hypothesis?
The first hit is often surgery, sepsis, trauma, hematologic malignancies, cardiac disease or massive transfusions. This causes neutrophils (PMNs) to sequester in the pulmonary capillaries. The second hit is thought to be the transfer of neutrophil, HLA antibodies, or bioactive lipids from donor blood product which activates the pulmonary PMNs. This activation results in endothelial damage → increased pulmonary capillary permeability → leakage of fluid and protein into the alveolar space.
How do you prevent TRALI?
The task of TRALI prevention mostly relies with the blood bank, not the clinicians! Unfortunately, there is nothing clinicians can do to mitigate the risk. Risk factors for TRALI are associated with donor selection and blood collection1. Female donors, specifically multiparous donors, increase risk of TRALI because they are more likely to develop HLA antibodies (formed to paternal HLA antigens in the fetus)3. TRALI is also more likely if donor is a close blood relative3.
- Delaney, M, et al. 2016. “Transfusion reactions: prevention, diagnosis, and treatment.” Lancet. 388: 2825-2836.
- Roubinian, NH, et al and the TRALI Study Group. 2017. “Differentiating pulmonary transfusion reactions using recipient and transfusion factors.” Transfusion. 57: 1684-1690.
- Looney, MR, Gropper, MA & Matthay, MA. 2004. “Transfusion-Related Acute Lung Injury.” CHEST. 126: 249-258.