Post-partum sequential occurrence of two diverse transfusion reactions (transfusion associated circulatory overload and transfusion related acute lung injury).

Transfusion associated circulatory overload (TACO) and transfusion related acute lung injury (TRALI) are two dissimilar pathological conditions associated with transfusion of blood products where the time course of the events and clinical presentation overlap leading to uncertainty in establishing the diagnosis and initiating the treatment, which otherwise differs. We encountered a case where a patient of post-partum hemorrhage developed TACO in the immediate post-operative period due to aggressive resuscitative attempts with blood products. The patient's condition was appropriately diagnosed and was managed according to the clinical scenario, and the condition abated. Subsequently, on the third post-operative day the patient again required blood product transfusions following which the patient developed TRALI, the diagnosis of which was also established and adequate treatment strategy was undertaken.

INTRODUCTION

Pulmonary edema with resultant hypoxemia occurring within 6 h of transfusion creates a diagnostic dilemma between transfusion associated circulatory overload (TACO) and transfusion related acute lung injury (TRALI). TACO is a form of hydrostatic edema, which occurs due to volume overload temporally associated with blood product transfusion. TRALI, on the other hand, is a form of permeability edema, which occurs due to leakage of fluid into the alveolar space due to diffuse alveolar capillary damage. We present a case wherein a patient of post-partum hemorrhage developed TACO in the immediate post-operative period following overzealous attempts at resuscitation with blood products. The same patient later developed TRALI after TACO had resolved. The course of events and management of the case are hereby described.

CASE REPORT

A 29-year-old female suffered post-partum hemorrhage following a normal vaginal delivery after, which she was referred to our institute on the second post-partum day. She had no co morbidities. On admission to the high dependency unit, she was in shock (blood pressure [BP] 80/50 mm Hg), had tachycardia (pulse rate 120/min), was restless and had severe pallor. Her baseline hemogram results were — hemoglobin 4 g/dl, Leucocyte count of 7800/ml and platelet count of 88 × 103/ml. Abdominal ultrasound was suggestive of placenta accreta. Immediate resuscitation commenced under anesthetist's supervision and packed red blood cells (PRBC's), fresh frozen plasma (FFP), platelet concentrates, and cryoprecipitate were arranged. Under ultrasound guidance and local anesthesia left internal jugular vein (for central line) and left radial artery (for arterial line) were cannulated. When conservative measures such as blood product transfusion (2 units PRBC and FFP each), uterine massage, administration of oxytocin, methylergometrine, prostaglandin and balloon tamponade failed to control the hemorrhage, the patient was rushed to the operation theater for an emergency hysterectomy. During the surgery, the patient lost around 1500 ml of blood for which she was transfused 3 units each of PRBC, FFP and platelets along with 1500 ml of hydroxyethyl starch over 3 h. Intraoperative central venous pressure (CVP) was maintained between 8 mm Hg and 10 mm Hg. She also required ionotropic support (dopamine at the rate of 8 μg/kg/min) transiently. At the conclusion of surgery, arterial blood gases (ABG) showed metabolic acidosis. (pH: 7.20, P/F ratio: 310, HCO3: 12, BE: −6.5), but ionotropic support could be tapered off. The patient was shifted to the intensive care unit (ICU) for elective mechanical ventilation. Within an hour of ICU admission, the patient developed an elevated CVP (14 mm Hg), jugular venous distension, and bilateral crepitations. Urgent work-up revealed hypoxemia (PO2: 70 mm Hg on FiO2 of 0.4) and cardiomegaly and lung congestion on chest X-ray [Figure 1]. Ultrasound chest showed B lines [Figure 2]. Electrocardiography and Troponin T levels were within normal limits. Echocardiography carried out promptly, showed a dilated left ventricle (left ventricular end diastolic diameter: 6 cm) with an ejection fraction of 50%. Plasma B-type natriuretic peptide (BNP) levels was measured, which was found elevated (255 pg/dL). A presumptive diagnosis of TACO was made and ventilator support was augmented by increasing the oxygen concentration (FiO2: 0.6), Positive End Expiratory Pressure (PEEP) (8 cm H2O), and pressure support (17 cm H2O). Diuretic therapy (furosemide 40 mg intravenous loading dose followed by 20 mg intravenously 8 h) was also initiated. Repeat chest X-ray after 24 h showed a reduction in cardiomegaly and clearing of congestion. Subsequently, her condition resolved and she could be extubated on the 2nd day of ICU stay. The next 2 days were uneventful with the patient receiving supportive therapy and making a gradual recovery. Her investigations and clinical features were within normal limits during this period. On the 3rd day from the event, oozing was observed from her central line insertion site, which prompted an emergency coagulogram. Her investigations revealed a platelet count of 115 × 103/ml, prothrombin time of 13 s, fibrinogen level of 252 mg/dl and an elevated activated partial thromboplastin time of 46 s. For correction of the same, she received 4 units of FFP over the next 5 h. Within 2 h of the last transfusion, she developed breathlessness, tachypnea and bilateral crepitations at the lung bases. At this point, the CVP measured was 5 mm Hg. Considering an allergic reaction, she was immediately administered steroids and antihistaminics, but was not relieved. An urgent chest X-ray revealed new onset bilateral pulmonary infiltrates [Figure 3]. ABG showed hypoxemia (PaO2 of 58.5 on FiO2-.35). A plasma sample for BNP was sent, which resulted in a value of 80 pg/dL. Echocardiography showed an ejection fraction of 55% without any valvular defects (left ventricular end diastolic diameter: 4.8 cm). Computed tomography angiography ruled out embolic phenomenon. As no other causes of acute lung injury (ALI) could be elicited, TRALI was considered a diagnosis of exclusion and the sample was sent back to the blood bank informing about the event with requisitioning for donor human leucocyte antigen (HLA) antibody testing. Ventilatory support was initiated (non-invasive ventilation) with the following settings: Mode Continous Positive Airway Pressure (CPAP), pressure support-16 cm H2O, PEEP-8 cm H2O and FiO2-0.5. After receiving ventilator support, the patient became comfortable. Repeat ABG's showed an improving trend and the ventilator support could be discontinued by the next 24 h. After this episode, further complications were not observed and the patient was discharged from the ICU on the 7th day of her surgery. A week later, the result of donor plasma came out to be positive for HLA antibody.

Figure 1

Chest X-ray showing cardiomegaly and pulmonary congestion

Figure 2

Ultrasound chest showing B Lines

Figure 3

Chest X-ray showing new onset infiltrates

DISCUSSION

TACO and TRALI lead to the development of similar conditions (pulmonary edema, hypoxemia and infiltrates on X-ray) in spite of their etiologies and pathophysiologies differing considerably. TACO has overall incidence of less than 1%.[1] Incidence rises in elderly and critically ill patients.[23] Clinically, TACO resembles hydrostatic pulmonary edema presenting with tachycardia, tachypnea, jugular venous distention, and elevated systolic BP.[14] Our patient was transfused approximately 900 ml of blood products during the initial 1 h of resuscitation and received another 1800 ml over a 3 h surgery. Post-operatively her signs of circulatory overload, the relevant imaging modalities and investigations suggested TACO as the diagnosis. TRALI is an immune reaction where leucocyte antibody binds to recipient's neutrophils activating them and causing endothelial damage in capillaries and lung. An alternate “two hit” theory is proposed where an initial insult causes endothelial activation and neutrophilic adhesion. A repeat insult activates neutrophils to release proteases and oxidases causing capillary leak and ALI. The diagnosis of TRALI is based on the clinical criteria based on National Heart, Lung and Blood Institute (NHLBI) working group and the Canadian Consensus Conference guidelines.[56] TRALI is a clinical diagnosis of exclusion mirroring respiratory distress following transfusion. Patients present with dyspnea, hypoxia, hypotension, pulmonary edema and pulmonary infiltrates emerging during or within 6 h of the completion of transfusion having no temporal relationship to an alternative lung injury factor. However, absence of jugular venous distention or S3 gallop rules out circulatory overload. TRALI lacks rapid relevant diagnostic tests. FFP has been implicated in most cases of TRALI and TRALI related deaths.[7] Since, we could not find any other plausible reason for developing ALI, based on the clinical course, time line of events, signs and relevant investigations, we diagnosed the second event as TRALI (or possible TRALI) and proceeded with its management. Amongst investigations, Chest X-rays are initial investigation to identify infiltrates and pulmonary edema, but have limited ability to identify the specific mechanism.[8] Echocardiography measures left ventricular ejection fraction; however, a normal value does not rule out hydrostatic edema due to diastolic dysfunction[9] as was observed in our patient during the first event. BNP is a neurohormone of cardiac origin secreted as a result of volume and pressure overload on the ventricles,[10] which provides a valuable adjunctive evidence in diagnosing TACO. An absolute BNP value of more than 100 pg/dl and pre-transfusion to the post-transfusion ratio of 1.5 is suggestive of TACO.[11] Laboratory tests, which strongly support, but are not required for clinical diagnosis of TRALI includes demonstration of HLA class I and Class II antibodies or neutrophil specific antibody in donor plasma and demonstration of corresponding antigen in donor neutrophils. Such investigation may take days and hence are not useful in acute clinical setting. Moreover, the mere demonstration of HLA and neutrophil antibody does not absolutely confirms the diagnosis of TRALI as not all donor antibodies result in TRALI.[12] Therefore, in our case, “possible TRALI” should be a better diagnosis.[12] Treatment of TACO and TRALI is supportive care. TACO resolved with ventilatory and diuretic therapy. TRALI was also treated supportively with ventilatory assistance. Diuretic usage here might have caused hypovolemia.[13]

CONCLUSION

Both TACO and TRALI are under recognized and underreported complications occurring due to blood product transfusion. It is difficult to differentiate between them especially in situations where both may co-exist. TACO is a mechanical event, which can be avoided by slow transfusion. TRALI necessitates work-up of donor for HLA antibodies and possible deferral. Use of plasma exclusively derived from male donors (multiparous women have higher levels of HLA antibodies), fresh platelets, and red cells and washed components are the other suggested strategies to reduce the risk of TRALI. However, none of these conditions is a contraindication for future transfusions. Awareness regarding the conditions, high clinical index of suspicion, proper clinical history and examination, and appropriate investigative tools are the cornerstone to diagnose and initiate successful management of these conditions.