HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response View responses Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (29) Citing Articles via Google Scholar Google Scholar Articles by Raghavan, M. Articles by Marik, P. E. Search for Related Content PubMed PubMed Citation Articles by Raghavan, M. Articles by Marik, P. E.

Anemia, Allogenic Blood Transfusion, and Immunomodulation in the Critically Ill^*

^* From the Department of Critical Care Medicine (Dr. Raghavan), University of Pittsburgh Medical Center, Pittsburgh; and Division of Pulmonary and Critical Care Medicine (Dr. Marik), Thomas Jefferson University, Philadelphia, PA.

Correspondence to: Paul Marik, MD, FCCP, Chief, Pulmonary and Critical Care Medicine, 1015 Chestnut St, Suite M100, Philadelphia, PA 19107; e-mail: paul.marik{at}jefferson.edu

	Abstract

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
Anemia and allogenic RBC transfusions are exceedingly common among critically ill patients. Multiple pathologic mechanisms contribute to the genesis of anemia in these patients. Emerging risks associated with allogenic RBC transfusions including the transmission of newer infectious agents and immune modulation predisposing the patient to infections requires reevaluation of current transfusion strategies. Recent data have suggested that a restrictive transfusion practice is associated with reduced morbidity and mortality during critical illness, with the possible exception of acute coronary syndromes. In this article, we review the immune-modulatory role of allogenic RBC transfusions in critically ill patients.

Key Words: blood transfusion • critically ill • ICU • immune modulation • infections • microchimerism • nosocomial infection

	Introduction

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
In recent years, blood transfusion requirements have been increasing due to the increasing burden of chronic disease in an aging population, improvement in life-support technology, increasing severity of illness in patients treated in the ICU, and other blood-intensive surgical procedures.¹² On the other hand, there is a trend toward decreasing blood donation and increasing cost due to the requirement for rigorous screening for transmittable infectious agents. In the United States alone, nearly 15 million U of blood are donated and 14 million U are transfused annually.² On average, 16% of patients in medical ICUs and 27% of those in surgical ICUs receive transfusions every day in the United States.³ In one series,⁴ 85% of patients with an ICU length of stay of > 1 week received at least 1 U of blood, with these patients receiving, on average, 9.5 U during their ICU stay. For decades, blood donation and transfusion were considered to be a life-saving strategy, and an arbitrary threshold of 10 g/dL was used as a transfusion trigger in critically ill patients.⁵ However, it has become evident that blood transfusion has immunomodulating effects that may increase the risk of nosocomial infections and cancer recurrence, and the possible development of autoimmune diseases later in life.⁶⁷⁸⁹¹⁰ Furthermore, the risk of "newer" transfusion-transmitted diseases has become recognized. Consequently, the safety of blood transfusions has been questioned and has led to a reevaluation of our blood transfusion practice.

	Anemia and Critical Illness

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
"Anemia of critical illness" is a common problem in the ICU. More than 90% of critically ill patients have subnormal hemoglobin levels by the third day of ICU admission.¹¹ In one series,¹² the mean hemoglobin level of patients admitted to the ICU was 11.3 g/dL, with 29% having a hemoglobin level of < 10 g/dL. Although anemia often results in extensive allogenic RBC transfusions among critically ill patients, there are insufficient data in the literature to support this widespread practice.

The etiology of anemia of critical illness is multifactorial and complex. Repeated phlebotomy procedures, GI blood loss, and other surgical procedures contribute significantly to the development of anemia.¹³¹⁴ Critically ill patients lose approximately 25 to 40 mL blood daily through phlebotomy, and patients with indwelling arterial catheters lose approximately 900 mL blood during their ICU stay.⁹¹⁵¹⁶ Other important contributing factors that exacerbate anemia in critically ill patients include coagulopathies, pathogen-associated hemolysis, hypoadrenalism, and nutritional deficiencies.¹⁷¹⁸¹⁹

RBC production in critically ill patients is often abnormal, and is involved in the development and maintenance of anemia. The pathophysiology of this anemia is complex, and includes the decreased production of erythropoietin (EPO), impaired bone marrow response to EPO, and reduced RBC survival.¹⁶ Critically ill patients have inappropriately low EPO concentrations, irrespective of the presence of acute renal failure.^2021222324 The suppression of EPO production by EPO gene inhibition²⁵ and EPO resistance are mediated by a variety of inflammatory mediators.²⁶ Interleukin (IL)-1, and tumor necrosis factor (TNF)- have been shown to inhibit EPO production.²⁷ Furthermore, IL-1, IL-6, and TNF- suppress erythropoiesis by direct inhibitory effects on bone marrow RBC production, while these effects can be reversed by exogenous EPO administration.²⁸

Decreased RBC synthesis and consequent anemia are also common during sepsis syndromes. Many ICU patients have low serum iron levels, total iron binding capacity, and elevated serum ferritin concentrations, suggesting the presence of "anemia of inflammation." Bacteria require iron for their growth, and several studies²⁹³⁰ have shown a link between iron and infection. It is therefore conceivable that the human host down-regulates iron metabolism and EPO synthesis as a component of nonspecific immunity during critical illness and sepsis. In addition, during sepsis low serum iron levels may also protect the host against iron-catalyzed oxidant cell damage.³¹ As RBCs also require iron for growth and maturation, anemia during sepsis may represent an adaptive mechanism by the host to starve the pathogen of iron. Thus, anemia of critical illness may also be viewed as "anemia of immune activation" and may have evolved as a protective mechanism against foreign antigens.

The most important physiologic consequence of anemia is a reduction in the oxygen-carrying capacity of blood. These changes are accompanied by increased cardiac output, a shift of the oxyhemoglobin dissociation curve, and increased oxygen extraction. RBC transfusions in the past have been routinely employed to augment tissue oxygen delivery. Although RBC transfusions increase systemic oxygen delivery, the immediate effectiveness of stored RBC transfusions to augment tissue oxygen uptake has been questioned in several studies.⁷³²³³ Furthermore, RBC transfusion has been associated with a higher incidence of postoperative infections and nosocomial ICU infections, and poorer outcome in critically ill patients.^34353637

	Trends in Transfusion Practice Among the Critically Ill

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
In a recent, large, multicentered observational study in the United States, Corwin et al³⁸ studied transfusion practices in 4,892 patients across 284 ICUs. Approximately 70% of patients who were admitted to the ICU had a baseline hemoglobin concentration of < 12 g/dL, and 44% of these patients received RBC transfusions. The mean (± SD) pretransfusion hemoglobin level was 8.6 ± 1.7 g/dL. The mortality rate was 10% for patients without transfusion, increasing to 25% for patients with 6 U transfused. Low hemoglobin levels were a common trigger for transfusion in approximately 90% of the patients, and the mean age of the RBCs transfused was 3 weeks; however, > 25% of transfused RBCs were > 1 month old.³⁸

In an earlier, large, multicentered, Canadian prospective trial published in 1999, Hebert et al³⁹ demonstrated that maintaining a hemoglobin level in the range of 7 to 9 g/dL was superior to a hemoglobin level of 10 g/dL, thereby raising questions regarding the validity of the historical assumption that RBC transfusions are beneficial for critically ill patients. The investigators enrolled 838 patients from 25 centers over a period of 3 years. Only normovolemic, anemic (plasma hemoglobin concentration, < 9 g/dL) patients who were expected to stay in the ICU for > 24 h were included in the study. Important exclusion criteria were evidence of active bleeding (ie, > 3 U transfused over 24 h), chronic anemia (plasma hemoglobin concentration < 9 g/dL in the preceding month), and cardiac surgery. The enlisted patients were randomized to receive either a restrictive transfusion strategy (hemoglobin concentration transfusion trigger, 7 g/dL; maintenance hemoglobin concentration range, 7 to 9 g/dL) or a liberal transfusion strategy (hemoglobin concentration transfusion trigger, 10 g/dL; maintenance hemoglobin concentration transfusion range, 10 to 12 g/dL). On average, a total of 2.6 U blood was administered to patients who were randomized to receive the restrictive strategy compared with 5.6 U for patients randomized to receive the liberal strategy. There was a nonsignificant trend toward decreased 30-day and 60-day all-cause mortality rates, and a lower adjusted multiple organ dysfunction score in favor of patients who were in the restrictive strategy group. There were significantly fewer cardiac complications, including acute myocardial infarction and pulmonary edema, observed in patients in the restrictive strategy group.

The investigators recommended a restrictive strategy as the best practice for most patients, including those with cardiovascular disease, but with the possible exception of critically ill patients with ongoing coronary ischemia. The appropriateness of a nonrestrictive transfusion approach for patients with ongoing coronary ischemia was supported by the publication of a later subgroup analysis⁴⁰ that suggested that patients with severe cardiac disease who had been randomized to the restrictive strategy group had a nonsignificant increase in the 30-day all-cause mortality rate. Unfortunately, the investigators excluded patients with chronic anemia and those undergoing cardiac surgery, and so it remains difficult to recommend a transfusion strategy for either of these groups on the basis of this study. Despite the study by Hebert et al,³⁹ Corwin et al³⁸ found that the transfusion practice in response to anemia has changed little in the United States in recent years. Since the study by Corwin et al³⁸ was initiated in early 2001(within 2 years of the study by Hebert et al³⁹), it is conceivable that by this time the universal implementation of a restricted transfusion strategy may not have occurred in all US ICUs.

Several studies⁸⁹³⁸⁴¹ have suggested that routine blood transfusions increase morbidity, mortality, and length of hospital stay in critically ill patients. In the study by Corwin et al,³⁸ patients receiving transfusions had more complications, including fever, fluid overload and hypotension, sepsis, thromboembolism, and ARDS. The number of units of blood transfused was independently associated with longer ICU and hospital length of stay and increased mortality.³⁸ However, this study included a heterogeneous group of critically ill patients and did not take into consideration the specific clinical scenarios during which patients received transfusions. Furthermore, this study does not answer the question of what is the appropriate pretransfusion hemoglobin level. Similar results were confirmed by Vincent et al⁹ in an earlier large, epidemiologic, observational study conducted among 146 European ICUs involving 3,534 patients. A total of 42% of patients received transfusions with an average pretransfusion hemoglobin level of 8.4 g/dL. Patients receiving transfusions had an average length of ICU stay of 7.2 days compared with 2.6 days for patients not receiving transfusions. Both the ICU and overall mortality rates (ICU mortality rate, 18.5% vs 10.1%, respectively; overall mortality rate, 29% vs 14.9%, respectively) were significantly higher for patients receiving transfusions than for patients not receiving transfusions. Patients receiving transfusions had higher rates of organ dysfunction and mortality for every hemoglobin level when compared to patients not receiving transfusions. Using propensity scores, the authors concluded that the associated risk of death was 33% for patients receiving transfusions compared with patients not receiving transfusions. However, the study had an observational design with no control for interventions, and included a wide variety of medical, surgical, and trauma patients, thereby confounding the interpretation of the results.⁹

	RBC Storage and Physiologic Alterations

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
RBCs undergo various morphologic and functional changes with storage, thereby mediating some of the adverse effects associated with allogenic transfusions in critically ill patients. RBCs stored for > 15 days have a decreased ability to deform and unload oxygen in the microcirculation.⁷ Complete depletion of 2,3-diphosphoglycerate concentrations occur after 2 weeks of storage, thereby reducing the ability of transfused RBCs to offload oxygen by > 50%.⁷⁴² RBC adenosine triphosphate levels have been shown to decrease following storage, resulting in a change in RBC shape from discoid to spherocytic, a loss of membrane lipid, and a decrease in cellular deformability.⁴³⁴⁴ This causes capillary sludging and obstruction, thereby predisposing the patient to tissue ischemia and decreased oxygen delivery.⁴⁵ The increased adhesion of nonleukodepleted stored RBCs to endothelial cells also has been demonstrated. RBC adhesion increases with the duration of storage, and prestorage leukoreduction eliminates such storage-related adhesion.⁴⁶ Therefore, the transfusion of adhesive RBCs may further compromise tissue blood flow, leading to impaired perfusion and organ dysfunction in critically ill patients.⁴⁷ The loss of endogenous RBC antioxidants occurs during the storage of blood. This increases oxidative injury of the cytoskeleton proteins and membrane phospholipids, and results in the conversion of hemoglobin to methemoglobin, which is incapable of binding oxygen.⁴⁸⁴⁹ The resultant tissue ischemia predisposes critically ill patients to an increased risk of infections and organ dysfunction.

WBCs are present in all cellular blood components that are prepared by standard techniques, and many studies have indicated that leukocyte contamination of erythrocyte or platelet preparations can cause a wide range of physiologic and immunologic dysfunction in recipients.^{10505152535455} The accumulation of various soluble bioactive substances occurs during storage, and includes histamine, lipids, cytokines, fragments of cellular membranes, soluble human leukocyte antigen (HLA) class I antigens, many of which are WBC-derived and play an important role in transfusion-induced immunomodulation (TRIM). Stored RBCs harbor potent proinflammatory cytokines such as IL-1, IL-6, IL-8, bactericidal permeability-increasing protein, and TNF.⁵⁶⁵⁷ The transfusion of stored RBCs has been shown to trigger neutrophil activation, and the release of IL-8 and secretory phospholipase A₂, thereby predisposing the patient to systemic inflammatory response syndrome.⁵⁸⁵⁹ The WBC contamination of stored RBC concentrates also has been shown to have a direct deleterious effect on RBC integrity. Increased hemolysis, microvesiculation, and potassium leakage occurs in RBCs in stored blood with an increasing amount of WBC contamination.⁶⁰ WBC apoptosis with a resultant release of toxic oxygen radicals and WBC-associated enzymes during RBC storage and transfusion have been implicated in some of these adverse effects.⁶¹ Arginase release from stored RBCs has been implicated in transfusion-associated immunosuppression. Arginine is degraded by arginase, an enzyme that is abundantly present in RBCs. While arginine stimulates lymphocyte function, arginase impairs it. Therefore, arginase leakage from stored RBCs may be an important mediator of immunosuppression that is associated with allogenic blood transfusions.⁶²

In the United States, approximately 20% of all RBCs transfused are 28 days old, and the RBC storage duration has been identified as a potential cause for the increased morbidity and mortality that has been observed with blood transfusions in several studies.^36376364 Clinical outcomes associated with increased storage duration include increased length of stay in the hospital/ICU, multiple organ system failure, increased infections, and impaired tissue oxygen utilization. Martin et al,⁶³ in one of the earliest retrospective analyses of 698 patients, described a relationship between the transfusion of non-leukocyte-reduced RBCs that had been stored for > 14 days and the associated increased length of stay in the ICU (p < 0.0001). Length of stay was significantly associated with the aging of RBCs (p = 0.003), the total number of units transfused (p = 0.004), and the median storage duration (p = 0.02). Furthermore, when transfused patients were analyzed separately from nontransfused patients, only RBC storage for > 14 days was independently predictive of length of stay (p < 0.0001).⁶³ Similarly, Purdy et al³⁶ described a positive correlation between mortality in patients with severe sepsis and the age of the non-leukocyte-reduced RBC units that were transfused. The median age of RBCs transfused to survivors was 17 days (range, 5 to 35 days) compared with 25 days (range, 9 to 36 days) for nonsurvivors (p < 0.0001).³⁶

Moore et al,⁶⁵ in a more recent prospective cohort study of 513 trauma patients, found that transfusion was an independent risk factor for postinjury multiple-organ failure, and described a clear dose-response relationship between the number of units transfused and the development of multiple-organ failure. Similarly Zallen and colleagues³⁷ demonstrated that in polytrauma patients the mean age of the blood, the number of units that were > 14 days old, and the number of units that were > 21 days old were all independent risk factors for multiple-organ failure.

However, two retrospective cohort studies⁶⁶⁶⁷ subsequently were unable to show an association between the age of stored transfused RBCs and postoperative length of stay in the hospital or ICU or with the duration of postoperative mechanical ventilation. The authors attributed this lack of effect to differences in the patient population, as both studies evaluated routine postoperative coronary artery bypass graft (CABG) surgery patients, whereas Martin et al⁶³ studied a heterogeneous group of critically ill patients, which excluded cardiac surgical patients.

More recently, Vamvakas and Carven⁶⁸ reexamined the CABG population, studying the effects of RBC supernatant, platelet supernatant, and plasma components on the duration of postoperative mechanical ventilation. Their results suggested an association between RBC supernatant volume and prolonged mechanical ventilation. Proinflammatory substances that accumulate during the storage of RBC concentrates were implicated in impairing pulmonary function in these patients.⁵⁷⁶⁸

In a study⁶⁷ of postoperative cardiac surgery patients, the transfusion of RBCs stored for > 28 days was an independent predictor of nosocomial pneumonia. Similarly, a prospective cohort study⁶⁹ of trauma patients demonstrated that the age of transfused blood was an independent risk factor for the development of major infections. The risk of major infection increased 13% for each unit that was > 14 days old, with the most common nosocomial infection being pneumonia.⁶⁹

	Immunology of Blood Transfusion

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
Allogenic blood transfusions introduce a multitude of foreign antigens including HLA class II-bearing donor dendritic antigen-presenting cells (APCs) in recipients.⁵³ The normal immune response to any foreign antigen is initiated by the recognition of foreign antigens associated with the major histocompatibility complex (MHC) by host T lymphocytes (Fig 1 ). After encountering the antigen, naive T cells receive the first signal through the T-cell receptor-MHC plus antigenic peptide complex and received the second signal through positive costimulatory molecules leading to full activation. Negative T-cell costimulatory pathways, on the other hand, tend to down-regulate immune responses.⁷⁰ Studies have indicated that costimulatory signals derived through non-MHC molecules that are present on APCs are required to elicit an immune response. Molecules such as B7–1 and B7–2 have been shown to provide critical early costimulatory signals through CD28 and CTLA-4 T-cell receptors, which regulate IL-2 secretion and clonal T-cell proliferation.⁷¹⁷² Various T-cell–antigen interactions induce the production of cytokines such as IL-2 and IL-4, which in turn activate T helper (Th) type 1 (IL-2) and Th-2 (IL-4) subsets, respectively. Th-2 in turn activates B-cell proliferation and antibody production. Thus, the immunogenicity of soluble, particulate, or cellular MHC antigens that are present on transfused allogenic blood products depend on the viability of APCs, the presence of costimulatory molecules to present them to recipient T cells, and HLA compatibility between donor and recipient. The impairment of any of these pathways, including the costimulatory molecules, has been shown to result in T-cell unresponsiveness.⁷⁰⁷²⁷³

View larger version (51K): [in this window] [in a new window] [Download PPT slide]   Figure 1. TRIM resulting from the interaction between donor APCs and recipient T cells. Recognition of donor HLA molecules as autologous antigens by recipient T cells results in immune tolerance and immunosuppression. Nonrecognition results in immune activation. Microchimerism results in immune suppression, allograft tolerance, GVHD, and autoimmune disease.

Although the exact mechanisms underlying TRIM still remains to be elucidated, allogenic blood transfusions have been shown to cause a decrease in the helper/suppressor T-lymphocyte ratio, a decrease in natural killer cell function, defective antigen presentation, the suppression of lymphocyte blastogenesis, and a reduction in delayed-type hypersensitivity and allograft tolerance.⁸⁵⁸⁶⁸⁷ Various bioactive soluble mediators are released from stored WBCs into human plasma during storage in a time-dependent manner as the WBCs deteriorate.⁸⁸ The concentrations of histamine, eosinophil cationic protein, eosinophilic protein X, myeloperoxidase, and plasminogen activator inhibitor-1 have all been reported to increase by threefold to 25-fold in the supernatant fluid of RBC components between day 0 and day 35 of storage.⁸⁹⁹⁰ Histamine, eosinophil cationic protein, and eosinophil protein X have been shown to inhibit neutrophil function, thereby contributing to the development of immune suppression and tissue damage.⁹¹⁹²

	Microchimerism

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
Microchimerism has been proposed as a possible mechanism of TRIM in allogenic transfusions.^5593949596 HLA compatibility between donor and recipient blood may result in the persistence of donor leukocytes and dendritic APCs within the recipient (microchimerism). It has been postulated that such chimerism may cause the down-regulation of the recipient’s immune response, including tolerance to donor alloantigens and allograft survival (Fig 1). Many years after pregnancy, liver transplantation, and neonatal exchange transfusions, microchimerism has been demonstrated, indicating a tolerance between donor and recipient cells. Microchimersim results in the release of IL-4, IL-10, and transforming growth factor-ß from Th-2 lymphocytes.⁹⁷ These cytokines have been shown to inhibit the production of Th-1 cells and proinflammatory cytokines, and to deactivate cytotoxic cells, thereby suppressing allograft rejection. Dendritic APCs also have been shown to cause recipient T-cell hyporeactivity, anergy, and depletion, thus mediating immunosuppression.⁹⁸ In such immunosuppressed recipients, microchimerism may result in the development of TAGVHD, polymorphous eruption of pregnancy, and other autoimmune connective tissue diseases such as scleroderma.^1099100 The loss of immunogenicity by transfused leukocytes in blood stored for > 2 weeks results in recipient T-cell anergy, thus potentiating immunodepression.¹⁰¹

	Deleterious Effects of RBC Transfusion in Critically Ill Patients

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
Transfusion Transmitted Infections
Transfusion-transmitted infections due to a variety of agents, although rare, remain a cause of concern in modern allogenic transfusion practice¹⁰²¹⁰³ (Tables 1, 2 ). Leukocyte contamination of blood products remains the primary etiologic mode of transmission of various infectious agents. Transfusion-transmitted cytomegalovirus (CMV) occurs in approximately 4% of transfusions and is due to the reactivation of latent CMV in leukocytes from healthy donors.¹⁰⁴¹⁰⁵ Besides CMV, other herpes viruses such as Epstein-Barr virus, human herpes virus-6, human herpes virus-7, and human herpes virus-8 are associated with leukocyte contamination during transfusion.⁵² Human T-cell leukemia/lymphoma virus (types I and II) targets T lymphocytes and is solely transmitted by cellular blood components.¹⁰⁶ Primary toxoplasmosis has been reported¹⁰⁷ to be transmitted by whole-blood, granulocyte transfusions and from transplantation of organs from seropositive donors to immunocompromised recipients. Although theoretical concerns exist regarding the possible transmission of Creutzfeldt-Jakob disease (CJD) and new-variant CJD by blood and leukocyte transfusion, newer epidemiologic studies¹⁰⁸ have failed to show a link between the transfusion and transmission of CJD. Transfusion-transmitted West Nile virus infection occurred in the United States in 2002 among 23 patients from 14 donors, and since then > 600 infected units of blood were identified from a 2.5-million donor pool.¹⁰²¹⁰⁹

Nosocomial Infections
Several studies^{86465115116117118119120} have clearly identified the increased risk of nosocomial infections among critically ill transfused patients. Currently, there are substantial data suggesting that exposure to allogenic leukocytes in transfusions may trigger an immune system response in the recipients, leading to an increased risk of infection, an earlier recurrence of malignancy, and an increased likelihood of mortality.¹⁴ Four possible mechanisms have been attributed to the development of bacterial infections following allogenic transfusions. These include the following: (1) a TRIM effect mediated by immunologically active allogenic WBCs that down-regulate the immune function of recipients^50101121; (2) a TRIM effect mediated by soluble biological response modifiers that are released in a time-dependent manner from WBC granules or membranes into the supernatant fluid of RBCs during storage⁹⁰¹²¹; (3) a TRIM effect mediated by soluble HLA peptides or other soluble mediators that circulate in allogenic plasma¹²²; and (4) a possible non-TRIM effect causing postoperative organ dysfunction that predisposes patients to infections.^57122123124

Nichols and colleagues¹¹⁵ published one of the first reports that linked transfusion with an increased incidence of infection in postoperative trauma patients with intestinal perforation and documented that the number of blood transfusions positively correlated with the postoperative infection rate. Similarly, Edna and Bjerkeset¹¹⁶ found an association between RBC transfusions and infectious complications in 484 patients with acute injuries. Infectious complications developed in 46 patients (9.5%). Logistic regression analysis revealed a relationship between blood transfusions and infectious morbidity that was independent of the other significant factors, including the injury severity score, age, and surgical procedure. The corrected odds ratios for infection were 1.6 (95% confidence interval [CI], 0.7 to 3.7) when 1 to 4 U blood were given and 6.4 (95% CI, 2.3 to 18.3) when > 4 U were used.¹¹⁶

Vamvakas and Carven⁶⁴ reported that colorectal surgery patients receiving perioperative allogenic blood transfusions have strikingly longer hospital stays than similar patients who do not receive transfusions. Length of stay increased by 1.3% (95% CI, 0.5 to 2.1%) per unit of RBCs and/or platelets transfused (p < 0.001), and hospital charges increased by 2.0% (95% CI, 1.4 to 2.6%) per unit (p < 0.001). In this study, allogenic transfusions were independently associated with longer hospital stays and higher hospital charges.⁶⁴ In two separate studies,¹²⁵¹²⁶ postoperative infection rates were similarly increased in transfused patients undergoing colorectal surgery secondary to trauma and cancer.

Braga and colleagues¹¹⁷ found that the transfusion of 1,000 mL of blood was an independent risk factor in the development of postoperative infection in patients undergoing operations for GI cancer. Ottino et al¹¹⁸ documented that RBC transfusion was an independent risk factor for sternal wound infection in 2,579 consecutive open-heart procedures. Similarly, several studies^35127128 have identified a higher risk for postoperative wound infections and pneumonia in patients undergoing CABG surgery following RBC transfusions in a dose-dependent manner. Patients with arm fractures or open leg fractures and burn patients have an increased risk of infection with transfusion when compared to nontransfused patients.¹¹⁹¹²⁰ Carson et al,¹²⁹ in a retrospective analysis, identified a 35% increased risk of nosocomial infections and a 52% increased risk of pneumonia in 9,598 hip fracture patients undergoing hip surgery who received allogenic RBC transfusions. Similarly, a linear trend between the number of units of RBCs transfused and the incidence of multiple-organ failure and infections also has been reported in trauma patients.^3765130

In critically ill patients, Taylor et al⁸ have demonstrated an association between RBC transfusion and nosocomial infection and mortality in a retrospective analysis of 1,717 patients. They investigated the rate of nosocomial infections in patients who had been admitted to a single medical-surgical-trauma unit over a period of approximately 2 years. A total of 416 patients (24%) received 1 U packed RBCs. The rate of nosocomial infection was strikingly higher and statistically significant in the group of patients who received transfusions compared with those who did not (15.4% vs 2.9%, respectively; p < 0.005). Moreover, a positive association was found between the number of transfusions and the incidence of nosocomial infections such that each unit of packed RBCs increased the risk of nosocomial infection by a factor of 1.5. Transfusion also was associated with both increased length of stay in the ICU and hospital and higher mortality rates. The relationship between transfusion and nosocomial infections persisted after the authors controlled for the probability of survival, age, and gender.

Shorr et al,¹³¹ in a secondary analysis of a large cohort of patients who received allogenic blood transfusions and mechanical ventilation, noted that transfusion independently increased the risk for ventilator-associated pneumonia. Of 4,892 subjects in the original cohort, 1,518 received mechanical ventilation for 48 h and did not have preexisting pneumonia. VAP was diagnosed in 20.5% of patients. The effect of transfusion on late-onset VAP was more pronounced (odds ratio, 2.16; 95% CI, 1.27 to 3.66) and demonstrated a positive dose-response relationship.¹³¹

Although the observational nature of these studies and the inability to control for all possible factors make it difficult to establish a cause-and-effect relationship and to separate the effects of transfusion from those of the underlying condition, the results of several prospective and randomized studies^132133134 have supported these findings. In these studies, the underlying hypothesis links the immunodepressant effect of transfusion to the presence of leukocytes (or leukocyte products).

These data have in turn led to the hypothesis that giving patients transfusions with leukocyte-reduced blood should result in reduced morbidity and mortality compared with patients receiving transfusions with non-leukocyte-reduced blood. However, most of the studies bearing on these questions have been flawed by retrospective design and inadequate consideration of the effects of comorbidities, whereas the few prospective studies in specific patient populations have reached contradictory conclusions. Metaanalyses of these substantial studies^134135136137 have failed to identify a statistically significant effect of leukocyte reduction. However, a recent study¹³⁸ evaluating clinical outcomes after the institution of a universal leukocyte reduction program in Canada noted a reduction in hospital mortality after the introduction of this program.

TRALI
TRALI is a life-threatening complication of allogenic transfusions and is the third most common cause of transfusion-associated death in the United States.¹³⁹ The estimated prevalence of TRALI is 1 in 1,120 cellular component transfusions with a mortality rate ranging from 1 to 10%.¹⁴⁰ Passively transferred donor blood containing antileukocyte antibodies (ie, IgG) directed against recipient leukocytes causes pulmonary sequestration, complement activation, and lung injury.¹⁴¹ In many cases, donor anti-HLA class II and antimonocyte antibodies are present.¹⁴²¹⁴³ Stored blood containing neutrophil-activating biological response modifiers such as bioactive lipids (lysophosphatidylcholines) and cytokines such as IL-6 and IL-8 predispose the patient to microcirculatory capillary lung injury.¹²³¹⁴⁰

TAGVHD
TAGVHD is a rare but lethal complication with a mortality rate > 90%, in which immunocompetent donor cells proliferate and attack host hemopoietic cells, skin, liver, and bile duct epithelial cells. Although more common in immunocompromised patients, the risk factors for the development of graft-vs-host disease (GVHD) include patients receiving transfusions from HLA-homozygous donors who are haploidentical with the patient, the use of relatives as donors, male recipients, and fresh blood containing viable lymphocytes.⁵⁴ Homozygosity for HLA occurs in 2% of the population, and the estimated chance of transfusion of an HLA homozygous product that is haploidentical with the patient is 1in 800 patients.¹⁴⁴ Thus, only a small fraction of such transfusions cause GVHD. However, it is now apparent that even histoincompatible WBCs circulate for up to a week in heavily transfused patients and thus have a potential to cause subclinical GVHD.¹⁴⁵ Irradiated and leukodepleted cellular products avert the development of GVHD.¹⁴⁶

Leukodepletion and Transfusion
Transfused leukocytes have been implicated in a variety of biological effects including febrile nonhemolytic transfusion reactions, transfusion-related alloimmunization to platelets, TRALI, and GVHD (Table 1). The potential beneficial effects of universal leukocyte depletion include a reduction in the incidence of nonhemolytic transfusion reactions, immunosuppression, and mortality.^{138147148149150} Both autologous and allogenic nonleukodepleted blood components release soluble bioactive mediators during storage, thereby mediating some of the TRIM effects, while the prestorage leukodepletion of allogenic cellular products has been shown to prevent some its deleterious effects.^{50518890134150} As prestorage leukodepletion is essential to prevent the accumulation of biological response modifiers, some studies comparing the incidence of postoperative infections.¹⁵¹ and cancer recurrence¹⁵² between the allogenic and autologous transfusions have reported similar outcomes, independent of the RBC component that was transfused.³⁵ It is also noteworthy that the negative effects of a liberal transfusion strategy observed in the Transfusion Requirements in Critical Care study³⁹ predated the implementation of universal leukoreduction in Canada. It is therefore conceivable that the detrimental effects of blood transfusion may be mediated primarily by the donor WBCs and their complex interactions with stored RBCs, rather than by the RBCs alone.

	Conclusion

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 
Concerns regarding the excess morbidity and mortality associated with nonrestrictive transfusion strategies, coupled with the emerging increased risk of the transmission of newer infectious agents, and immunomodulation, should prompt the reevaluation of current transfusion protocols in critically ill patients. A restrictive transfusion strategy appears to improve outcomes in critically ill patients. Lowering the trigger for transfusion to a hemoglobin concentration of approximately 7 g/dL in patients without coronary disease and implementing other blood conservation techniques, such as minimizing phlebotomy, the reuse of discarded blood by using closed circuits of blood sampling, and the use of recombinant EPO, should help to lower transfusion requirements. In addition, the implementation of prestorage leukodepleted blood, along with pathogen inactivation techniques may reduce the adverse effects associated with allogenic transfusions. The benefit of fresh leukodepleted blood (ie, < 15 days) compared to leukodepleted old blood (ie, > 15 days) has yet to be determined. However, the major clinical dilemma is not between the use of fresh vs old blood and/or leukocyte-depleted vs non-leukocyte-depleted blood but between stored blood vs no blood.

	Footnotes

 
Abbreviations: APC = antigen-presenting cell; CABG = coronary artery bypass graft; CI = confidence interval; CJD = Creutzfeldt-Jakob disease; CMV = cytomegalovirus; EPO = erythropoietin; GVHD = graft-vs-host disease; HLA = human leukocyte antigen; IL = interleukin; MHC = major histocompatibility complex; SEN-V = SEN virus; TAGVHD = transfusion-associated graft-vs-host disease; Th = T helper; TNF = tumor necrosis factor; TRALI = transfusion-related lung injury; TRIM = transfusion-induced immunomodulation; TTV = TT virus

Received for publication February 20, 2004. Accepted for publication August 12, 2004.

	References

TOP Abstract Introduction Anemia and Critical Illness Trends in Transfusion Practice... RBC Storage and Physiologic... Immunology of Blood Transfusion Microchimerism Deleterious Effects of RBC... Conclusion References

 

1. Wells, AW, Mounter, PJ, Chapman, CE, et al (2002) Where does blood go? Prospective observational study of red cell transfusion in north England. BMJ 325,803[Abstract/Free Full Text]
2. National Blood Data Resource Center. Comprehensive report on blood collection and transfusion in the United States. 2001 National Blood Data Resource Center. Bethesda, MD:
3. Groeger, JS, Guntupalli, KK, Strosberg, M, et al Descriptive analysis of critical care units in the United States: patient characteristics and intensive care unit utilization. Crit Care Med 1993;21,279-291[ISI][Medline]
4. Corwin, HL Blood transfusion in the critically ill patient. Dis Mon 1999;45,409-426[CrossRef][Medline]
5. Cane, RD Hemoglobin: how much is enough? Crit Care Med 1990;18,1046-1047[ISI][Medline]
6. Hebert, PC, Fergusson, DA Red blood cell transfusions in critically ill patients. JAMA 2002;288,1525-1526[Free Full Text]
7. Marik, PE, Sibbald, WJ Effect of stored-blood transfusion on oxygen delivery in patients with sepsis. JAMA 1993;269,3024-3029[Abstract]
8. Taylor, RW, Manganaro, L, O’Brien, J, et al Impact of allogenic packed red blood cell transfusion on nosocomial infection rates in the critically ill patient. Crit Care Med 2002;30,2249-2254[CrossRef][ISI][Medline]
9. Vincent, JL, Baron, JF, Reinhart, K, et al Anemia and blood transfusion in critically ill patients. JAMA 2002;288,1499-1507[Abstract/Free Full Text]
10. Martin, L, Watier, H, Vaillant, L, et al Sjogren’s syndrome and vitiligo in a woman with posttransfusion microchimerism. Am J Med 2001;111,419-421[CrossRef][ISI][Medline]
11. Corwin, HL, Rodriguez, RM, Pearl, RG, et al Erythropoietin response in critically ill patients [abstract]. Crit Care Med 1997;25,A82
12. von Ahsen, N, Muller, C, Serke, S, et al Important role of nondiagnostic blood loss and blunted erythropoietic response in the anemia of medical intensive care patients. Crit Care Med 1999;27,2630-2639[CrossRef][ISI][Medline]
13. Brown, RB, Klar, J, Teres, D, et al Prospective study of clinical bleeding in intensive care unit patients. Crit Care Med 1988;16,1171-1176[ISI][Medline]
14. Corwin, HL, Parsonnet, KC, Gettinger, A RBC transfusion in the ICU: is there a reason? Chest 1995;108,767-771[ISI][Medline]
15. Smoller, BR, Kruskall, MS Phlebotomy for diagnostic laboratory tests in adults: pattern of use and effect on transfusion requirements. N Engl J Med 1986;314,1233-1235[Abstract]
16. van Iperen, CE, Gaillard, CA, Kraaijenhagen, RJ, et al Response of erythropoiesis and iron metabolism to recombinant human erythropoietin in intensive care unit patients. Crit Care Med 2000;28,2773-2778[ISI][Medline]
17. Batge, B, Filejski, W, Kurowski, V, et al Clostridial sepsis with massive intravascular hemolysis: rapid diagnosis and successful treatment. Intensive Care Med 1992;18,488-490[CrossRef][ISI][Medline]
18. Campillo, B, Zittoun, J, de Gialluly, E Prophylaxis of folate deficiency in acutely ill patients: results of a randomized clinical trial. Intensive Care Med 1988;14,640-645[ISI][Medline]
19. Rodriguez, RM, Corwin, HL, Gettinger, A, et al Nutritional deficiencies and blunted erythropoietin response as causes of the anemia of critical illness. J Crit Care 2001;16,36-41[CrossRef][ISI][Medline]
20. Corwin, HL, Gettinger, A, Pearl, RG, et al Efficacy of recombinant human erythropoietin in critically ill patients: a randomized controlled trial. JAMA 2002;288,2827-2835[Abstract/Free Full Text]
21. Elliot, JM, Virankabutra, T, Jones, S, et al Erythropoietin mimics the acute phase response in critical illness. Crit Care 2003;7,R35-R40[CrossRef][ISI][Medline]
22. Krafte-Jacobs, B, Levetown, ML, Bray, GL, et al Erythropoietin response to critical illness. Crit Care Med 1994;22,821-826[ISI][Medline]
23. Krafte-Jacobs, B Anemia of critical illness and erythropoietin deficiency. Intensive Care Med 1997;23,137-138[CrossRef][ISI][Medline]
24. Rogiers, P, Zhang, H, Leeman, M, et al Erythropoietin response is blunted in critically ill patients. Intensive Care Med 1997;23,159-162[CrossRef][ISI][Medline]
25. Frede, S, Fandrey, J, Pagel, H, et al Erythropoietin gene expression is suppressed after lipopolysaccharide or interleukin-1 beta injections in rats. Am J Physiol 1997;273,R1067-R1071[ISI][Medline]
26. Jelkmann, WE, Fandrey, J, Frede, S, et al Inhibition of erythropoietin production by cytokines: implications for the anemia involved in inflammatory states. Ann N Y Acad Sci 1994;718,300-309[Abstract]
27. Johnson, CS, Keckler, DJ, Topper, MI, et al In vivo hematopoietic effects of recombinant interleukin-1 in mice: stimulation of granulocytic, monocytic, megakaryocytic, and early erythroid progenitors, suppression of late-stage erythropoiesis, and reversal of erythroid suppression with erythropoietin. Blood 1989;73,678-683[Abstract/Free Full Text]
28. Johnson, CS, Cook, CA, Furmanski, P In vivo suppression of erythropoiesis by tumor necrosis factor-alpha (TNF-): reversal with exogenous erythropoietin (EPO). Exp Hematol 1990;18,109-113[ISI][Medline]
29. Fishbane, S Review of issues relating to iron and infection. Am J Kidney Dis 1999;34,S47-S52[ISI][Medline]
30. Jurado, RL Iron, infections, and anemia of inflammation. Clin Infect Dis 1997;25,888-895[ISI][Medline]
31. Balla, G, Vercellotti, GM, Muller-Eberhard, U, et al Exposure of endothelial cells to free heme potentiates damage mediated by granulocytes and toxic oxygen species. Lab Invest 1991;64,648-655[ISI][Medline]
32. Fernandes, CJ, Jr, Akamine, N, De Marco, FV, et al Red blood cell transfusion does not increase oxygen consumption in critically ill septic patients. Crit Care 2001;5,362-367[CrossRef][ISI][Medline]
33. Gramm, J, Smith, S, Gamelli, RL, et al Effect of transfusion on oxygen transport in critically ill patients. Shock 1996;5,190-193[ISI][Medline]
34. Mynster, T, Nielsen, HJ The impact of storage time of transfused blood on postoperative infectious complications in rectal cancer surgery: Danish RANX05 Colorectal Cancer Study Group. Scand J Gastroenterol 2000;35,212-217[CrossRef][ISI][Medline]
35. Vamvakas, EC, Carven, JH Transfusion and postoperative pneumonia in coronary artery bypass graft surgery: effect of the length of storage of transfused red cells. Transfusion 1999;39,701-710[CrossRef][ISI][Medline]
36. Purdy, FR, Tweeddale, MG, Merrick, PM Association of mortality with age of blood transfused in septic ICU patients. Can J Anaesth 1997;44,1256-1261[Abstract/Free Full Text]
37. Zallen, G, Offner, PJ, Moore, EE, et al Age of transfused blood is an independent risk factor for postinjury multiple organ failure. Am J Surg 1999;178,570-572[CrossRef][ISI][Medline]
38. Corwin, HL, Gettinger, A, Pearl, RG, et al The CRIT Study: anemia and blood transfusion in the critically ill; current clinical practice in the United States. Crit Care Med 2004;32,39-52[CrossRef][ISI][Medline]
39. Hebert, PC, Wells, G, Blajchman, MA, et al A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care: Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340,409-417[Abstract/Free Full Text]
40. Hebert, PC, Yetisir, E, Martin, C, et al Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001;29,227-234[CrossRef][ISI][Medline]
41. Hebert, PC, Wells, G, Marshall, J, et al Transfusion requirements in critical care: a pilot study; Canadian Critical Care Trials Group. JAMA 1995;273,1439-1444[Abstract]
42. Apstein, CS, Dennis, RC, Briggs, L, et al Effect of erythrocyte storage and oxyhemoglobin affinity changes on cardiac function. Am J Physiol 1985;248,H508-H515[Medline]
43. Fischer, DJ, Torrence, NJ, Sprung, RJ, et al Determination of erythrocyte deformability and its correlation to cellular ATP release using microbore tubing with diameters that approximate resistance vessels in vivo. Analyst 2003;128,1163-1168[CrossRef][Medline]
44. Stuart, J, Nash, GB Red cell deformability and haematological disorders. Blood Rev 1990;4,141-147[CrossRef][ISI][Medline]
45. Simchon, S, Jan, KM, Chien, S Influence of reduced red cell deformability on regional blood flow. Am J Physiol 1987;253,H898-H903[Medline]
46. Luk, CS, Gray-Statchuk, LA, Cepinkas, G, et al WBC reduction reduces storage-associated RBC adhesion to human vascular endothelial cells under conditions of continuous flow in vitro. Transfusion 2003;43,151-156[CrossRef][ISI][Medline]
47. Ho, J, Milkovic, S, Gray, L, et al Transfusion of stored red blood cells (RBC) occlude the rat microvaculature in-vivo. Blood 2001;98,544a
48. Wolfe, LC, Byrne, AM, Lux, SE Molecular defect in the membrane skeleton of blood bank-stored red cells: abnormal spectrin-protein 4.1-actin complex formation. J Clin Invest 1986;78,1681-1686[ISI][Medline]
49. Racek, J, Herynkova, R, Holecek, V, et al Influence of antioxidants on the quality of stored blood. Vox Sang 1997;72,16-19[CrossRef][ISI][Medline]
50. Bordin, JO, Heddle, NM, Blajchman, MA Biologic effects of leukocytes present in transfused cellular blood products. Blood 1994;84,1703-1721[Abstract/Free Full Text]
51. Jensen, LS, Kissmeyer-Nielsen, P, Wolff, B, et al Randomised comparison of leucocyte-depleted versus buffy-coat-poor blood transfusion and complications after colorectal surgery. Lancet 1996;348,841-845[CrossRef][ISI][Medline]
52. Chu, RW Leukocytes in blood transfusion: adverse effects and their prevention. Hong Kong Med J 1999;5,280-284[Medline]
53. Austyn, JM Antigen uptake and presentation by dendritic leukocytes. Semin Immunol 1992;4,227-236[Medline]
54. Brand, A Immunological aspects of blood transfusions. Transpl Immunol 2002;10,183-190[CrossRef][ISI][Medline]
55. Dzik, WH Mononuclear cell microchimerism and the immunomodulatory effect of transfusion. Transfusion 1994;34,1007-1012[CrossRef][ISI][Medline]
56. Stack, G, Baril, L, Napychank, P, et al Cytokine generation in stored, white cell-reduced, and bacterially contaminated units of red cells. Transfusion 1995;35,199-203[CrossRef][ISI][Medline]
57. Fransen, E, Maessen, J, Dentener, M, et al Impact of blood transfusions on inflammatory mediator release in patients undergoing cardiac surgery. Chest 1999;116,1233-1239[CrossRef][ISI][Medline]
58. Mynster, T, Dybkjoer, E, Kronborg, G, et al Immunomodulating effect of blood transfusion: is storage time important? Vox Sang 1998;74,176-181[CrossRef][ISI][Medline]
59. Zallen, G, Moore, EE, Ciesla, DJ, et al Stored red blood cells selectively activate human neutrophils to release IL-8 and secretory PLA2. Shock 2000;13,29-33[ISI][Medline]
60. Hogman, CF, Meryman, HT Storage parameters affecting red blood cell survival and function after transfusion. Transfus Med Rev 1999;13,275-296[CrossRef][ISI][Medline]
61. d’Almeida, MS, Jagger, J, Duggan, M, et al A comparison of biochemical and functional alterations of rat and human erythrocytes stored in CPDA-1 for 29 days: implications for animal models of transfusion. Transfus Med 2000;10,291-303[CrossRef][ISI][Medline]
62. Prins, HA, Houdijk, AP, Nijveldt, RJ, et al Arginase release from red blood cells: possible link in transfusion induced immune suppression? Shock 2001;16,113-115[ISI][Medline]
63. Martin, CM, Sibbald, WJ, Lu, X, et al Age of transfused red blood cells is associated with ICU length of stay. Clin Invest Med 1994;17,B21a
64. Vamvakas, EC, Carven, JH Allogeneic blood transfusion, hospital charges, and length of hospitalization: a study of 487 consecutive patients undergoing colorectal cancer resection. Arch Pathol Lab Med 1998;122,145-151[ISI][Medline]
65. Moore, FA, Moore, EE, Sauaia, A Blood transfusion: an independent risk factor for postinjury multiple organ failure. Arch Surg 1997;132,620-624[Abstract]
66. Vamvakas, EC, Carven, JH Length of storage of transfused red cells and postoperative morbidity in patients undergoing coronary artery bypass graft surgery. Transfusion 2000;40,101-109[CrossRef][ISI][Medline]
67. Leal-Noval, SR, Jara-Lopez, I, Garcia-Garmendia, JL, et al Influence of erythrocyte concentrate storage time on postsurgical morbidity in cardiac surgery patients. Anesthesiology 2003;98,815-822[CrossRef][ISI][Medline]
68. Vamvakas, EC, Carven, JH Allogeneic blood transfusion and postoperative duration of mechanical ventilation: effects of red cell supernatant, platelet supernatant, plasma components and total transfused fluid. Vox Sang 2002;82,141-149[CrossRef][ISI][Medline]
69. Offner, PJ, Moore, EE, Biffl, WL, et al Increased rate of infection associated with tr