Secondary Logo

Journal Logo

Are multiple blood transfusions really a cause of acute respiratory distress syndrome?

Sadis, C.*; Dubois, M.-J.*; Mélot, C.*; Lambermont, M.; Vincent, J.-L.*

European Journal of Anaesthesiology: April 2007 - Volume 24 - Issue 4 - p 355–361
doi: 10.1017/S0265021506001608

Background and objectives: Multiple blood transfusions are considered a common cause of acute respiratory distress syndrome (ARDS). We hypothesized that ARDS is more a consequence of ARDS risk factors (in particular circulatory shock) requiring transfusions than a result of the transfusions themselves.

Methods: This retrospective study included 103 patients admitted during a 10-month period to an 858-bed university hospital who received multiple transfusions (more than six units of packed red blood cells in 24 h).

Results: Ten patients developed ARDS; they were more commonly admitted with circulatory shock (36 (38.7%) vs. 8 (80%), P = 0.01), polytrauma (7 (7.5%) vs. 4 (40%), P = 0.01) or thoracic trauma (3 (3.2%) vs. 4 (40%), P = 0.01). The sequential organ-failure assessment (SOFA) score at admission was higher in patients who developed ARDS than in those who did not (9.0 ± 3.1 vs. 5.6 ± 3.4, P < 0.005). The total amount of transfusion in the first 24 h was 14.0 ± 6.8 U in the ARDS patients and 10.6 ± 7.3 U in the other patients (P = 0.17); the differences remained non-significant in the following days. During the first 24 h, patients who developed ARDS received more fresh frozen plasma than those who did not (21.8 ± 10.6 U vs. 10.7 ± 14.7 U, P = 0.02). Patients who developed ARDS had lower PaO2/FiO2 ratios (114 ± 61 mmHg vs. 276 ± 108 mmHg, P = 0.01), lower arterial pH (7.27 ± 0.10 vs. 7.34 ± 0.11, P = 0.06) and higher minute volume (10.6 ± 2.8 L min1 vs. 7.9 ± 1.8 L min1, P = 0.03) than patients without ARDS. Multivariable analysis retained thoracic trauma and hypoxaemia during the first 24 h (but not multiple transfusions) as independent risk factors for ARDS.

Conclusions: In this retrospective study, the development of ARDS in massively transfused patients was less related to poly-transfusion than to other factors related to circulatory shock, polytrauma or thoracic trauma. Thoracic trauma and a low PaO2 during the first 24 h were identified as independent risk factors for ARDS.

Free University of Brussels, Erasme Hospital, *Department of Intensive Care, Belgium

Blood Transfusion Service, Belgium

Correspondence to: Jean-Louis Vincent, Department of Intensive Care, Erasme University Hospital, Route de Lennik 808, B-1070 Brussels, Belgium. E-mail:; Tel: +32 2 555 3380; Fax: +32 2 555 4555

Accepted for publication 15 October 2006

First published online 7 November 2006

Back to Top | Article Outline


Transfusions are usually listed among the causes of acute lung injury (ALI) or acute respiratory distress syndrome (ARDS) [1-4] although there is little evidence in the literature for a cause and effect relationship. There may be some confusion between so-called transfusion-related acute lung injury (TRALI) and other forms of ALI/ARDS. Although both are due to an inflammatory reaction resulting in an increase in the permeability of the alveolo-capillary membrane with extravasation of inflammatory cells and plasma [1,5,6], TRALI generally occurs early after transfusion and is usually transient [7]. Massive transfusions are typically required in the context of hypovolaemia, abnormal homeostasis and altered tissue oxygenation, so that these factors may be more directly related to the pathophysiology of respiratory failure. Other factors like chest trauma, cardiopulmonary bypass (CPB) or pancreatitis may also be significantly involved in the development of ALI/ARDS [1,2].

Therefore, we hypothesized that the development of ALI/ARDS following multiple transfusions (massive or repetitive) is the consequence of shock or other factors requiring transfusion rather than the transfusion itself.

Back to Top | Article Outline

Patients and methods

We reviewed Blood Bank Records of all patients hospitalized in the 858-bed Erasme University Hospital during a 10-month period (September 2000-June 2001) and identified 103 adult patients who had received at least six units of red blood cells in a 24-h period. We recorded whether the patient had been admitted to the intensive care unit (ICU), the diagnosis at admission and the main risk factors for ARDS other than transfusion (circulatory shock, sepsis, CPB, polytrauma, thoracic trauma, pneumonia, inhalation, lung transplantation, acute pancreatitis, fat embolus and drowning) [1,2], the length of stay in the ICU for those poly-transfused patients who were admitted to the ICU and the outcome.

The day of the transfusion was considered as D0. Data about additional transfusions, cardiovascular, respiratory, renal, haemostatic and hepatic functions and blood lactate concentration were recorded for the four subsequent days (D1, D2, D3, D4). The diagnostic criteria for ALI were hypoxaemia (PaO2/FiO2 ≤ 300 mmHg) of acute onset without evidence of left ventricular dysfunction (when measured, the pulmonary-artery wedge pressure was <18 mmHg) with bilateral pulmonary infiltrates on the chest X-ray [8]. The diagnostic criteria for ARDS were identical except for a PaO2/FiO2 ≤ 200 mmHg [8]. Circulatory shock was defined as severe hypotension (mean arterial pressure <70 mmHg) associated with signs of altered tissue perfusion and increased blood lactate concentrations, requiring the administration of vasopressor catecholamines. Organ dysfunction at admission was assessed using the sequential organ-failure assessment (SOFA) score [9].

Statistical analysis included χ2-tests on discrete variables and t-tests for continuous variables with results expressed in mean ± standard deviation (SD) in all cases; P < 0.05 was considered as statistically significant. Logistic regression was used to identify independent variables linked to ARDS and to mortality with a value of P < 0.05 taken as the limit of statistical significance.

Back to Top | Article Outline


Of the 103 poly-transfused patients, 87 (84.4%) were admitted to the ICU for an average length of ICU stay after the transfusion of 7.6 days (range 1-50 days). The main causes of admission were bleeding during cardiovascular surgery, gastrointestinal bleeding, transplantation (hepatic (16), lung (three), cardiac (two) and renal (two)) or polytrauma (Table 1). The other causes of admission were urological surgery (six patients), neurosurgery (three patients), orthopaedic surgery (one patient), thyroidectomy (one patient) and severe anaemia in a context of myelodysplasia (one patient). ARDS risk factors present at admission included circulatory shock, CPB, polytrauma, thoracic trauma and pneumonia (Table 1).

Table 1

Table 1

Back to Top | Article Outline

Development of ARDS

ARDS developed in 10 patients (9.7%), with a mean age of 52 ± 20 yr (range 19-77 yr), who were treated in the ICU for 20.4 ± 16.4 days. Initial diagnoses were polytrauma (four, including three with thoracic trauma), major cardiovascular surgery (four) or gastrointestinal bleeding (two) (Table 2). On admission, eight patients had a circulatory shock of 2.4 ± 2.0 days duration (1-6 days).

Table 2

Table 2

The average time to develop ARDS after the first day of massive transfusion was 1.5 ± 0.7 days (1-3 days); the ARDS criteria were present for 8.4 ± 7.2 days (3-21 days). The SOFA score at admission was significantly higher in patients who developed ARDS than in those who did not (9.0 ± 3.1 vs. 5.6 ± 3.4, P < 0.005) (Fig. 1). Patients who developed ARDS more commonly had circulatory shock, polytrauma, thoracic trauma or pneumonia than those who did not (Table 3). Patients who developed ARDS also had a longer length of stay in ICU than the other patients (20.4 ± 16.4 vs. 9.4 ± 15.7 days, P = 0.04) (Table 3).

Figure 1.

Figure 1.

Table 3

Table 3

During the first 24 h after massive transfusion and on subsequent days, there was no significant difference in total transfusion between patients with and without ARDS (14.0 ± 6.8 U vs. 10.6 ± 7.3 U, P = 0.17). The amount of transfused leucodepleted packed red blood cells was identical in patients with and without ARDS (3.0 ± 4.3 U vs. 3.1 ± 4.1 U P = 0.94). However, the amount of fresh frozen plasma (FFP) given during the first 24 h was significantly larger in patients who developed ARDS (21.8 ± 10.6 U vs. 10.7 ± 14.7 U, P = 0.02) (Table 4).

Table 4

Table 4

During the first 24 h after massive transfusion, patients who developed ARDS had a lower PaO2, a higher FiO2, a lower PaO2/FiO2 ratio and a lower pH than the other patients (Table 4). The minute ventilation during the first 24 h after transfusion was significantly higher in patients who developed ARDS (10.6 ± 2.8 L min1) than in those who did not (7.9 ± 1.8 L min1) (P = 0.03) but the differences in tidal volume, PaCO2 and respiratory rate during the first 24 h did not reach statistical significance (Table 4).

Using a logistic regression technique, thoracic trauma and a low PaO2 during the first 24 h were identified as independent risk factors for ARDS.

Back to Top | Article Outline


Seventy-six patients (73.8%), including 5 of the 10 patients with ARDS, survived. Factors associated with mortality were circulatory shock at admission, in particular if prolonged, gastrointestinal bleeding and pneumonia (Table 3). Bleeding during a transplantation procedure had a better prognosis as 22 of 23 transplanted patients survived.

During the first 24 h, non-survivors had a higher FiO2, a lower pH, a lower PaCO2 and a higher arterial lactate concentration than the survivors (Table 4). Non-survivors were also more often treated by mechanical ventilation during the first 24 h. More non-survivors received vasopressor catecholamines during the first 24 h (Table 4).

There was no significant difference between survivors and non-survivors in the amount of leucodepleted blood received (3.2 ± 4.3 U vs. 2.9 ± 3.7 U, P = 0.77).

Logistic regression analysis showed that gastrointestinal bleeding and a high arterial lactate concentration during the first 24 h were independent variables related to death.

Back to Top | Article Outline


ARDS is a clinical syndrome developing as a consequence of a systemic reaction associated with the release of inflammatory mediators. Many factors can contribute to its development, and blood transfusions are usually listed among them [1-4]. However, patients requiring multiple transfusions usually have other factors that could favour the development of ARDS, including circulatory shock, polytrauma and chest trauma [1,2]. In the present study, multiple transfusion, as defined by the transfusion of at least six units of packed red blood cells during 24 h, was associated with a relatively low incidence of ARDS (only 10%). A multivariable analysis revealed that the development of ARDS was related more to underlying factors than to the multiple transfusions.

Considerable effort has been deployed to avoid complications related to transfusion of blood products, including acute haemolysis subsequent to ABO incompatibility, other immunological alterations and transmission of infectious micro-organisms [10-12]. Specific respiratory alterations following transfusions were first recognized in 1951 by Barnard [13], but it was only in 1983 that Popovsky and colleagues [14] introduced the term TRALI. TRALI is attributed to a leucoagglutination phenomenon and to the presence of lymphotoxic antibodies targeted against the white blood cells of the transfused patient [15-17]. Lipids released in the plasma during red blood cell conservation may also be involved [18].

Clinically, symptoms of TRALI (including dyspnoea, cyanosis, arterial hypotension, fever and cough) typically occur 1-2 h after transfusion and persist for 4-6 h after the transfusion [15,16]. Data on the frequency of TRALI are rare and vary widely as definitions of TRALI differ considerably among studies, but a recent consensus panel summarized an incidence rate of between 1 in 4000 and 1 in 557 000 per unit of red blood cells transfused, 1 in 432 and 1 in 88 000 per unit of platelets transfused, and 1 in 8000 and 1 in 74 000 per unit of FFP transfused [6]. Even though TRALI often requires mechanical ventilation, the associated mortality is not as high as that associated with other causes of ARDS (6% vs. 30-50%) [19]. The incidence of TRALI has probably decreased over time, especially with the implementation of routine leucodepletion, which may reduce the incidence of immunological reactions during transfusion [20,21]. In our study, no difference in the development of ARDS or mortality was attributed to leucodepletion, but the size of the study limits the interpretation of this finding.

All blood components have been incriminated in the occurrence of TRALI [22-24]. In our study, only the amount of FFP given during the first 24 h was higher for patients who developed ARDS, but this may reflect the greater severity of the disease state in these patients, even though one study incriminated FFP as the leading cause of ARDS after pneumectomy [25]. During the first 24 h, ARDS patients receive more FFP either because their bleeding cannot be controlled or because they develop coagulopathy. In these situations, haemorrhagic shock occurs or will occur [26].

In practice, it is difficult to separate TRALI from other causes of ALI/ARDS. In our study, patients who developed ARDS already had more severe hypoxaemia in the first 24 h. On the other hand, circulatory shock, polytrauma and thoracic trauma, which are considered as risk factors for ARDS [1,2], were more common in the ARDS patients. Massive transfusions are usually required in the context of circulatory shock and it seems that shock, more than transfusion, was the cause of ARDS. In our study, polytrauma was a common cause of ARDS, and is associated with the systemic release of inflammatory mediators [27]. Polytrauma, in particular when it includes a thoracic component, contributes to a pulmonary inflammatory reaction and to the development of ARDS. A higher SOFA score in the patients who developed ARDS compared with those who did not indicated a more severe degree of organ failure at admission. During the first 24 h, patients with ARDS also required more vasopressor agents and had somewhat more severe lactic acidosis than the other patients. Likewise, Eberhard and colleagues showed that the degree of initial metabolic acidosis was a reliable predictor for ALI in trauma patients [28]. Another study on polytrauma patients [29] showed that arterial lactate concentration is a good predictor of early ARDS and the related liberation of the cytokines.

Silliman and colleagues have already noted that TRALI only develops if immunological reactions in the blood are associated with specific clinical situations [18,30]. Clinical situations necessitating massive transfusion include shock, polytrauma and thoracic trauma. In our study, patients who developed ARDS required a higher minute volume ventilation (due to a combination of a higher tidal volume and faster respiratory rate) than the other patients. This was likely to compensate for a more severe metabolic acidosis. Although large tidal volumes are associated with a worse outcome in ALI/ARDS, and may be associated with a greater inflammatory response that could contribute to lethality even in the early stages of the acute illness [31], tidal volume was not retained in the multivariable model.

Mortality was associated with both shock and lactic acidosis. Patients with gastrointestinal bleeding had the worst prognosis, while transplantation was associated with a good prognosis, probably because transplantations are performed in patients who are kept in an optimal state before the intervention and because bleeding could be controlled.

Our study is limited by its retrospective nature and the limited number of patients. Massive transfusions are relatively rare over 10 months in an 850-bed university hospital, even though the criterion we used for massive transfusion, six transfused units in 24 h, may be somewhat lower than that used in some other studies [32,33]. This relatively low definition for massive transfusion may in itself be seen as a limitation; however, Gong and colleagues [34] reported a stepwise increase in the rate of lung injury in those receiving just one to three units of red blood cells, suggesting that the requirement for massive transfusion as a risk factor for ALI may be overly restrictive [35]. The differences in the patients' predisposing factors for ARDS and in their source of bleeding may also potentially have influenced the results.

In conclusion, our study suggests that the development of ARDS following massive transfusions may be due to factors other than massive transfusion per se.

Back to Top | Article Outline


1. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342: 1334-1349.
2. Weinacker AB, Vaszar LT. Acute respiratory distress syndrome: physiology and new management strategies. Annu Rev Med 2001; 52: 221-237.
3. Milot J, Perron J, Lacasse Y et al. Incidence and predictors of ARDS after cardiac surgery. Chest 2001; 119: 884-888.
    4. Miller PR, Croce MA, Kilgo PD, Scott J, Fabian TC. Acute respiratory distress syndrome in blunt trauma: identification of independent risk factors. Am Surg 2002; 68: 845-850.
    5. Dry SM, Bechard KM, Milford EL, Churchill WH, Benjamin RJ. The pathology of transfusion-related acute lung injury. Am J Clin Pathol 1999; 112: 216-221.
    6. Kleinman S, Caulfield T, Chan P et al. Toward an understanding of transfusion-related acute lung injury: statement of a consensus panel. Transfusion 2004; 44: 1774-1789.
    7. Toy P, Popovsky MA, Abraham E et al. Transfusion-related acute lung injury: definition and review. Crit Care Med 2005; 33: 721-726.
    8. Bernard GR, Artigas A, Brigham KL et al. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. The American-European Consensus Conference on ARDS. Am J Respir Crit Care Med 1994; 149: 818-824.
    9. Vincent JL, Moreno R, Takala J et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Med 1996; 22: 707-710.
    10. Sazama K. Reports of 355 transfusion-associated deaths: 1976 through 1985. Transfusion 1990; 30: 583-590.
    11. AuBuchon JP, Kruskall MS. Transfusion safety: realigning efforts with risks. Transfusion 1997; 37: 1211-1216.
      12. 1997 Committee on Transfusion Medicine of the American Society of Anesthesiologists. Adverse effects of transfusion. Questions and Answers about Transfusion Practices. Park Ridge: American Society of Anesthesiologists, 2002: 16-19.
      13. Barnard RD. Indiscriminate transfusion: a critique case report illustrating hypersensitivity reactions. N Y State J Med 1951; 51: 2399-2402.
      14. Popovsky MA, Abel MD, Moore SB. Transfusion-related acute lung injury associated with passive transfer of antileucocyte antibodies. Am Rev Respir Dis 1983; 128: 185-189.
      15. Kopko PM, Popovsky MA. Pulmonary injury from transfusion-related acute lung injury. Clin Chest Med 2004; 25: 105-111.
      16. Kopko PM, Marshall CS, MacKenzie MR, Holland PV, Popovsky MA. Transfusion-related acute lung injury: report of a clinical look-back investigation. JAMA 2002; 287: 1968-1971.
      17. Looney MR, Gropper MA, Matthay MA. Transfusion-related acute lung injury: a review. Chest 2004; 126: 249-258.
      18. Silliman CC, Voelkel NF, Allard JD et al. Plasma and lipids from stored packed red blood cells cause acute lung injury in an animal model. J Clin Invest 1998; 101: 1458-1467.
      19. Popovsky MA, Moore SB. Diagnostic and pathogenetic considerations in transfusion-related acute lung injury. Transfusion 1985; 25: 573-577.
      20. Hebert PC, Fergusson D, Blajchman MA et al. Clinical outcomes following institution of the Canadian universal leucoreduction program for red blood cell transfusions. JAMA 2003; 289: 1941-1949.
      21. Yazer MH, Podlosky L, Clarke G, Nahirniak SM. The effect of prestorage WBC reduction on the rates of febrile nonhemolytic transfusion reactions to platelet concentrates and RBC. Transfusion 2004; 44: 10-15.
      22. Kawamata M, Miyabe M, Omote K, Sumita S, Namiki A. Acute pulmonary edema associated with transfusion of packed red blood cells. Intensive Care Med 1995; 21: 443-446.
      23. Leger R, Palm S, Wulf H, Vosberg A, Neppert J. Transfusion-related lung injury with leucopenic reaction caused by fresh frozen plasma containing anti-NB1. Anesthesiology 1999; 91: 1529-1532.
        24. Virchis AE, Patel RK, Contreras M et al. Lesson of the week. Acute non-cardiogenic lung oedema after platelet transfusion. BMJ 1997; 314: 880-882.
        25. van der Werff YD, van der Houwen HK, Heijmans PJ et al. Postpneumonectomy pulmonary edema. A retrospective analysis of incidence and possible risk factors. Chest 1997; 111: 1278-1284.
        26. Stainsby D, MacLennan S, Hamilton PJ. Management of massive blood loss: a template guideline. Br J Anaesth 2000; 85: 487-491.
        27. Dunham CM, Frankenfield D, Belzberg H et al. Inflammatory markers: superior predictors of adverse outcome in blunt trauma patients? Crit Care Med 1994; 22: 667-672.
        28. Eberhard LW, Morabito DJ, Matthay MA et al. Initial severity of metabolic acidosis predicts the development of acute lung injury in severely traumatized patients. Crit Care Med 2000; 28: 125-131.
        29. Rixen D, Siegel JH. Metabolic correlates of oxygen debt predict post trauma early acute respiratory distress syndrome and the related cytokine response. J Trauma 2000; 49: 392-403.
        30. Silliman CC, Paterson AJ, Dickey WO et al. The association of biologically active lipids with the development of transfusion-related acute lung injury: a retrospective study. Transfusion 1997; 37: 719-726.
        31. Su F, Nguyen ND, Creteur J et al. Use of low tidal volume in septic shock may decrease severity of subsequent acute lung injury. Shock 2004; 22: 145-150.
        32. Donaldson MD, Seaman MJ, Park GR. Massive blood transfusion. Br J Anaesth 1992; 69: 621-630.
        33. Hewitt PE, Machin SJ. ABC of transfusion. Massive blood transfusion. BMJ 1990; 300: 107-109.
        34. Gong MN, Thompson BT, Williams P et al. Clinical predictors of and mortality in acute respiratory distress syndrome: potential role of red cell transfusion. Crit Care Med 2005; 33: 1191-1198.
        35. Nathens AB. Massive transfusion as a risk factor for acute lung injury: association or causation? Crit Care Med 2006; 34: S144-S150.


        © 2007 European Society of Anaesthesiology