Transfusion‐associated circulatory overload: the plot thickens

Physicians have known for decades that transfusion is a risk factor for fluid overload-cardiogenic pulmonary edema but the perception was that compared to other complications, transfusion-associated circulatory overload (TACO) was unimportant. We now know otherwise. TACO was “rediscovered” once transfusion-related acute lung injury (TRALI) was placed on the transfusionists' map. As TRALI gained acceptance for its morbidity and frequency, TACO was increasingly identified as the “other” pulmonary complication. What was viewed as simply an outcome of too much blood given too quickly is now understood to be a more subtle problem that can be difficult to diagnose. Contrary to prior belief, TACO is a common complication of transfusion, but the frequency is not firmly established. Like TRALI, the reported incidence varies widely. In two multicenter analyses of patients undergoing total hip or total knee replacement surgery, TACO was observed in 1 and 8 percent of cases, respectively.1-3 In the latter study involving 9482 patients, the TACO group received 1.8 to 2.7 units of RBC per patient. In a Mayo Clinic study from the 1980s, 20 percent of the identified cases involved only 1 unit, contradicting the perception that large volumes of red blood cells (RBCs) are essential to trigger the reaction.4 Robillard and colleagues5 recently demonstrated that TACO is not limited to the very young or elderly. In a report from the Quebec hemovigilance system, 14 percent of cases involved patients 18 to 59 years of age and 19 percent were 60 to 69 years old. However, there are populations that appear to be at increased risk. In a study of intensive care patients who did not require respiratory support at the time of transfusion, Rana and coworkers6 found that 25 of 49 patients with confirmed acute pulmonary edema were found to have TACO. The incidence of TACO was 1 in 356 per unit transfused, compared to 1 in 534 per unit for TRALI. TACO has not been taken seriously due to the perception that its clinical impact was inconsequential. Supplemental oxygen, a dose of diuretics—problem solved! But the facts don't bear this out. In the Massachusetts Medicare study, Popovsky and colleagues1 demonstrated that patients with TACO required more intensive care and a longer length of stay. Food and Drug Administration (FDA) data demonstrate that TACO is the third most common cause of death from transfusion.7 From fiscal year 2005 to 2006, fatal case reports increased from 2 to 13 percent of all fatalities. In the French hemovigilance system, over a 6-year period 742 cases were identified, resulting in 27 deaths (3.6% of all fatalities).8 As the United States moves to its own biovigilance program, undoubtedly more fatalities will be identified. TACO should be relatively straightforward to diagnose. In the setting of acute respiratory distress, key findings include posttransfusion hypertension, widened pulse pressure, tachycardia, and elevated central venous and pulmonary wedge pressure. Pulmonary edema and cardiomegaly are seen on chest radiographs. Confusion arises with the overlap between TRALI and TACO, particularly in the critical care setting. In TRALI, up to 15 percent of cases present with hypertension. In both entities, hypoxemia is prominent and the timelines for presentation of the signs and symptoms are both within 2 hours of transfusion. At the Mayo Clinic, 80 percent of initial TRALI reports are subsequently diagnosed as TACO (S.B. Moore, personal communication, 2005). A recent study intended to detect TRALI by computer-based screening found that of 88 cases of posttransfusion hypoxemia, 10 of these were TACO.9 These observations suggest that relying on physician reports to determine the incidence of TACO will result in underreporting; it is unlikely that many physicians who recognize that their medical practice precipitated pulmonary edema in a patient who received transfusion will be eager to report their misadventures to the transfusion service. With the development of assays for B-natriuretic peptide (BNP), there was optimism that the quantitative test would be able to differentiate TACO from TRALI. BNP is a 32-amino-acid peptide that is a cardiac-derived neurohormone. It is synthesized and secreted from the ventricular myocardium in response to ventricular volume and pressure distension.10 Originally approved by the FDA to aid the diagnosis of congestive heart failure and acute coronary syndromes, several investigators have evaluated the usefulness of this test for pulmonary transfusion reactions. In a case-control study involving patients diagnosed with TACO, Zhou and colleagues11 found that a posttransfusion to pretransfusion BNP ratio of at least 1.5 was associated with a sensitivity of 81 and specificity of 89 percent. Using a more stable marker, NT-BNP, Tobian and coworkers12 reported comparable sensitivity and specificity. In both studies, the control groups were transfused patients with or without other types of transfusion reactions, with no evidence of TRALI. In this issue of TRANSFUSION, Li and colleagues13 expand our understanding of natriuretic peptides with a prospective cohort study in which BNP and NT-pro-BNP were used to correlate with the diagnosis of TACO, TRALI, and possible TRALI. Critical care experts blinded to the NP levels determined the diagnoses using consensus conference definitions. One-hundred fifteen adult patients from intensive care units were enrolled if they presented with acute pulmonary edema within 6 hours of transfusion. Patients with pulmonary edema before transfusion as well as those who died within 6 hours after transfusion were excluded. The median time from recognition of pulmonary edema to the time of testing was 21 hours for BNP and 12 hours for NT-pro-BNP. The results yielded two important findings. The first is that of the 115 patients, 50 (43%) had TACO, demonstrating just how prevalent TACO is in the intensive care setting. Both BNP and NT-pro-BNP levels were significantly higher in patients with TACO compared to those with TRALI and possible TRALI. When the authors evaluated the diagnostic performance of the test (i.e., its ability to discriminate these conditions) the positive predictive value of these tests was only in the range of 74 to 78 percent. The lack of discrimination of these tests may be a function of the stimulatory effect of hypoxemia on BNP and NT-pro-BNP secretion.14 In TACO and TRALI, right heart dysfunction and hypoxemia are the norm. Thus, although our enthusiasm for natriuretic peptides may be tempered by these results, the accuracy of the peptide assays could have been affected by the timing of the sample collection and other comorbidities. The results may also vary outside the intensive care setting. A surprising finding in the study by Li and coworkers was that the calculated fluid balance in the TACO patients was lower compared to those with TRALI, despite higher cardiac filling pressures in the former. Other investigators have reported the opposite—a positive fluid balance.1 The explanation for the Mayo findings are unclear but it is difficult to accurately measure fluid balance in general. Most practitioners are in agreement that “rapid infusion” is a contributing factor for circulatory overload, so it is ironic that there are few studies as to what constitutes an appropriate infusion rate for blood components, particularly RBCs. While the AABB Technical Manual recommends an infusion rate of 150 to 300 mL per hour for RBCs and faster rates for plasma and platelets, the science behind these recommendations is sparse.15 Almost certainly, recipient weight should factor into an infusion equation. The lack of evidence-based guidelines is compounded by poor control of infusion rates in the clinical setting, attributable to inconsistent use of infusion pumps and insufficient bedside monitoring of the transfusion process. In one unpublished report, 47 cases of TACO in adults were associated with infusion rates of 0.9 to 48.1 mL per minute (C. Andrzejewski, personal communication, 2005). As with TRALI, early recognition should lead to better clinical outcomes. The bedside nurse is the key to this change. Andrzejewski and colleagues16 found at 15 minutes after transfusion, systolic blood pressure, pulse pressure, and mean arterial pressure were significantly higher in fluid-challenged patients than those not overloaded. The search for bedside heralds that identify the patient at risk for TACO and other complications must continue.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.