Transfusion Related Acute Lung Injury
Richard Charleswood
Outline
Transfusion Related Acute Lung Injury
(TRALI)
Richard Charlewood
New Zealand Blood Service
Some History
• 1901 ABO discovered
• 1940 Rh described
• 1943 Transfusion transmitted hepatitis first recognised
• 1951 TRALI first described described
• 1985 TRALI first case series
• 2004 Consensus conference
Rare?
• The leading cause of transfusion-related deaths in USA
• The 2nd commonest in the UK after ABO-incompatible transfusion in the UK
(SHOT)
• Incidence varies widely and with components transfused
– Platelets: 1 in 432 to 1 in 88,000
– Red cells: 1 in 4,000 to 1 in 557,000
– FFP: 1: 7,900 to 1 in 74,000
• Difficulty with standard definition, methods of surveillance and denominators
Presentation
• SOB, acute hypoxia, fever, tachypnoea, tachycardia, hypotension unresponsive
to fluids
• Copious frothy sputum
• CXR: bilateral alveolar & interstitial infiltrates, leading to white-out
• Normal pulmonary wedge pressure / JVP
ie Non-cardiogenic pulmonary oedema indistinguishable from ARDS
Within 6 hours of transfusion
– most within 1-2 hours
At autopsy
• Features are consistent with ARDS
– widespread leukocyte infiltration
– interstitial and inter-alveolar oedema
– hyaline membrane formation
– destruction of normal lung parenchyma
Risk factors
• Seen in all ages, male = female
• History of transfusion reactions not predictive
• Seen especially after major surgery, trauma, severe infection
• ?also more common in
• TTP, cardiac disease, haem malignancy
• Orthotopic liver transplants
• Seen mainly after whole blood, red cells, platelets, FFP
• Usually >60ml plasma needed
• Rarely after cryo, IVIG and stem cells
Consensus diagnostic criteria
a. Acute Lung Injury (ALI)
i. Acute onset
ii. Hypoxia - PaO2/FiO2 ? 300 mmHg2
- or O2 sats < 90% on RA
- or other clinical evidence of
hypoxia
iii. Bilateral infiltrates on frontal CXR
iv. No evidence of circulatory overload
b. No pre-existing ALI pre-transfusion
c. During or within 6 hours of transfusion
d. No temporal relationship to alternative risk factor for ALI
Possible TRALI
a. Acute Lung Injury (ALI)
b. No pre-existing ALI pre-transfusion
c. During or within 6 hours of transfusion
d. A clear temporal relationship to alternative risk factor for ALI
Risk factors for ALI
• Direct lung injury: Aspiration, pneumonia, toxic inhalation, lung contusion,
near drowning
• Indirect lung injury: Severe sepsis, shock, multiple trauma, burn injury,
acute pancreatitis, cardiopulmonary bypass, drug overdose
Differential Diagnosis
• ARDS
– same clinical features
• TACO or fluid overload
– tachypnoea, tachycardia, hypertension, cyanosis
• Anaphylactic transfusion reaction
– bronchospasm, wheeze, tachypnoea, hypotension, cyanosis, erythematous rash with urticaria
• Bacterial transfusion transmitted infection
– fever, hypotension, respiratory distress, abrupt onset
• Early acute haemolytic transfusion reaction
– haemolysis usually obvious
Investigations
• Clinical diagnosis
• Meet diagnostic criteria
– Chest X-ray, O2 sats or blood gas, no circulatory overload
• If intubated, can measure protein in oedema fluid
• high in TRALI and ARDS (exudate)
• low in cardiogenic pulmonary oedema (transudate)
• ?role of Brain Natriuretic Peptide (BNP)
– high in cardiac overload
– ?favours TRALI if low
Pathogenesis
Two theories
• Immunological
– Antibodies to leukocytes
• Non-immunological
– Biologically active lipids
• Two theories are not necessarily mutually exclusive
Immunological theory
• First hypothesis
• Antibodies to HLA and Neutrophil antigens
• Antibody from donor
– typically a multiparous female
– antibodies made to partner’s HLA/HNA antigens expressed on circulating fetal white cells
• Occasionally antibody from recipient to donor wbc
– Smaller pool of wbc to react with if leukodepleted
Immunologic mechanism
Antibody vs wbc ? C' activation -> pulmonary leukostasis & wbc activation
or
antibody activates neutrophil via Fc receptors -> release of inflammatory mediators, vascular permeability & capillary leak
Monocytes and endothelial cells may also be targeted
Problems with Immunologic theory
• Antibody found in up to 89%, but not found in up to 15%
• ?Antibody not detected because
• antibody of unknown specificity
• antibody sequestered but recipient leukocytes
• HLA antibody but low titre or not C' binding
Non-immunologic mechanism
• Similar to ARDS in animal models
1. Clinical condition
• Major surgery, trauma, severe infection
• massive transfusion, haematological malignancy
• TTP, solid organ transplants, warfarin reversal with FFP
• cardiovascular disease needing bypass surgery
2. Infusion of a biologic response modifier
• biologically active lipids (cell fragments)
• cytokines (IL6, TNF-a, IL8)
• leukoagglutinating antibodies
• Retrospective study (10 TRALI cases)
– All 10 had a “first event” clinical condition
– Only 20% of controls (FNHTR cases)
• Rat lungs exposed to lipopolysaccharide followed by
– platelet lipid extracts or
– Day 5 platelet supernatant-> pulmonary oedema (compared
with saline)
– No change with prestorage leukodepletion
• Can't explain all cases - no response modifier in FFP
Treatment
Supportive
• Maintain haemodynamics
• Aggressive lung-protective respiratory support
– but less than 70% will need ventilation
• Vasopressors where needed
• No diuretics unless also fluid overloaded
• No steroids
• ?? others
• Prostaglandin E1
• Anti-endotoxin Ab
• NSAID
• Anti-TNF-alpha
• Pentoxyphilline
• Surfactant
Prognosis
• 80% improve within 48 - 96 hours (unlike ARDS)
• CXR resolution within 1 - 4 days
BUT
• 20% still hypoxic with infiltrates at day 7
• 5-10% mortality
Investigation
• All allogeneic blood components received within 6 hours of TRALI
• Anti HLA class I & II antibodies
• Anti neutrophil antibodies (HNA)
• Crossmatch samples with HLA/HNA antibodies against recipient white cells
Or
• Determine specificity and/or cross-reactivity against recipient’s tissue type
Interpretation
• If concordance, = TRALI
• antibody to corresponding antigen
• Lymphocyte crossmatch positive
• If no cross-reactivity, may still be TRALI (40%)
• specificity of antibody
• If no antibody, does not exclude TRALI
Future transfusions
• Antibody may be from recipient
• 10-15% possibility in un-leukodepleted blood
• lower in leukodepleted
• Therefore a risk of recurrence
• Monitor closely during & after subsequent transfusions
• Stop if dyspnoea, frothy sputum
• Leukodepletion
• (already in place in NZ)
Prevention of TRALI by Blood Service
A number of options:
• Defer all multiparous donors
– HLA antibodies in
• 7.8% nullips
• 19.2% primips
• 26-50% of multips
– Study comparing multip plasma vs control in ICU
• ?pO2 sats, ?TNFa with multip plasma
• ?AP with control, not multips
– But major impact on donor supply
• ? of all apheresis donors in USA
• Defer all implicated donors permanently
– current practice in NZ
– a lookback study on donor with anti-5b
• 36% of 36 charts evaluated = reactions
– 8 severe reactions in 8 patients
• only 2 considered TRALI at the time
– 7 mild-moderate in 6 patients
– a lookback study on donor high titre leukocytotoxic HLA-I antibody
• 15% of donations had a transfusion reaction
• Use only male plasma for FFP
– already implemented in UK
– early results look promising
– some implication for blood supply
– doesn’t address platelets and red cells
– being piloted in New Zealand
• Platelet additive solution
– improves viability of platelets in storage
– decreases amount of plasma in unit
– won’t necessarily decrease cell fragments
• Prestorage leukodepletion
– Associated with less complement activation
– Already implemented in New Zealand
• Discourage maternal directed donations
– Higher risk for concordant HLA antibodies
• Increase level of awareness of disorder
• Co-ordinated reporting
– National Haemovigilance Programme
Prevention of TRALI at the bedside
Transfuse less, increased level of awareness
• RCT in critically ill patients
– restrictive (70g/l) vs liberal transfusion (10g/l) threshold
– restrictive = less ALI (7.7% vs 11.4%)
– ?some of ALI = TRALI
• 150,000 patients with ALI in USA pa
– one third exposed to multiple blood products at time of onset
– ?association causal or reflects underlying illness
|
Ministry
of Health
