Healthy individuals are able to tolerate profound, short-term decreases in hemoglobin levels and oxygen saturation without serious consequences, but critically ill patients in respiratory failure lack the necessary reserve capacity to preserve tissue oxygenation. The development of progressive anemia in ICU patients has led to much interest and debate about transfusion practices, yet optimal hemoglobin levels and how they should be achieved remain unclear. Animal and human studies regarding critical oxygen delivery provide the rationale for optimizing hemoglobin levels and supporting cardiovascular function during respiratory failure. Theoretically, the oxygen-carrying benefit of RBCs should hasten recovery from respiratory failure, and transfusions would therefore be expected to shorten the duration of mechanical ventilation. However, evidence to the contrary has been reported. Controversies related to transfusions and their inability to improve outcomes suggest that further research regarding transfusion alternatives is needed, especially in anemic patients with respiratory failure.
Key words: anemia; oxygen delivery; respiratory failure; transfusion; intensive care
Abbreviations: CABG = coronary artery bypass graft: CO = cardiac output; Dos = oxygen delivery; Sa[O.sub.2] = arterial oxygen saturation; V[O.sub.2] = oxygen consumption
Learning Objectives: 1. To review the basic determinants and alterations observed in oxygen delivery, during respiratory failure. 2. To summarize clinical studies that have examined thctors such as anemia and transfusions and their effect on outcomes in patients with respiratory failure.
**********
Normal homeostasis requires the interdependent actions of the respiratory and circulatory systems to provide a flow of oxygenated blood to the tissues in order to maintain aerobic metabolism. (1) Respiratory failure, circulatory collapse, or profound decrements in the oxygen-carrying capacity of the blood may lead to life-threatening consequences. Moderate degrees of dysfunction in the respiratory, circulatory, or hematologic systems may not be life threatening if there is physiologic reserve capacity in the other two limbs of this aerobic support system triad. (2) Survival of patients with respiratory failure requires not only treatment of the primary condition but also attention to deficits or potential deficits in relevant associated organ systems. Anemia may have serious consequences in the patient with respiratory failure if oxygen delivery (D[O.sub.2]) to the tissues is impaired and aerobic metabolism is not maintained. (3-5)
DETERMINANTS OF D[O.sub.2]
Tissue hypoxia in the critically ill patient is a serious problem that may result from either defects in D[O.sub.2] or impaired oxygen utilization at the cellular level. The principal determinants of D[O.sub.2] are arterial oxygen saturation (Sa[O.sub.2]), hemoglobin levels, and cardiac output (CO). (6) It is important to note that these factors are weighted equally in the equations traditionally used to describe their relationships (Table 1). In reality, however, variations in hemoglobin and CO are often much more profound in clinical contexts than are variations in Sa[O.sub.2]. For example, it is relatively common to see critically ill patients with hemoglobin levels that are one half to two thirds of normal, or similar variations in CO, whereas it is unusual to see sustained reductions of Sa[O.sub.2]. Conceptually, this means that the management of the patient with respiratory failure should include attention to the optimization of the hemoglobin levels and the cardiovascular system if one is to ensure that critically ill patients maintain adequate D[O.sub.2]. The utility of this approach will be subsequently discussed.
NORMAL PHYSIOLOGIC RESERVE
Normal, healthy human beings have a significant physiologic reserve capacity and are able to tolerate substantial decrements in Sa[O.sub.2] and hemoglobin without ill effects and to significantly augment CO in response to stressors. Experiments in normal volunteers simulating exposure to hypoxic conditions seen at the summit of Mount Everest (8,840 m, barometric pressure of 240 mm Hg, inspiratory oxygen pressure of 43 mm Hg) demonstrated that a very low Sa[O.sub.2] could be well tolerated. (7) Before exercise, resting individuals were found to have an average Sa[O.sub.2] of 59%. Compensation for the low Sa[O.sub.2] occurred with an increase in the CO to an average of 8.02 L/min. This occurred mainly in association with an increase in the average heart rate to 104 beats/min in normal subjects. With these changes, the average calculated D[O.sub.2] is 936 mL/min, which is normal. Healthy persons can also tolerate very low hemoglobin levels. Using techniques of isovolemic hemodilution, hemoglobin can be reduced to 5 g/dL in normal volunteers without evidence of decreased oxygen consumption (V[O.sub.2]). (8) Again, a robust cardiovascular response manifested by increases in heart rate and stroke volume leads to compensation. Exercise in normal, healthy people leads to significant increases in not only V[O.sub.2] but also D[O.sub.2] via augmentation of CO and oxygen extraction. (9) In critical illness, anemia and an impaired cardiovascular system are potential factors limiting the response to hypoxia in respiratory failure.
CRITICAL ILLNESS: D[O.sub.2] AND V[O.sub.2]
While the associations of Sa[O.sub.2], hemoglobin levels, and CO with D[O.sub.2] are well defined, the relationship of the reduction in Doe to the development of tissue hypoxia in the critically ill is complex and not completely understood. Compensatory mechanisms that could account for tolerance of varying D[O.sub.2] at the tissue or cellular level might include increased oxygen extraction from the blood or increased efficiency of oxygen utilization.
The relationship between V[O.sub.2] and D[O.sub.2] has been studied in both animal models and humans. In 1977, Cain (10) published the results of a classic experiment in which 27 dogs were subjected to conditions involving isovolemic hemodilution, [beta]-blockade, and administration of low oxygen gas mixtures. These experiments revealed that a single biphasic curve described the relationship between D[O.sub.2] and V[O.sub.2] (Fig 1). (1) Above the value termed the critical threshold for D[O.sub.2], V[O.sub.2] was independent of its delivery. Below this threshold value, V[O.sub.2] decreased in a linear fashion as D[O.sub.2] decreased. The D[O.sub.2] critical threshold in these animal experiments was 9.8 mL/kg/min.
[FIGURE 1 OMITTED]
The existence of this biphasic relationship and the presence of a D[O.sub.2] critical threshold has important implications in the management of seriously ill patients, including those in respiratory failure. One potential implication of this theory is that modulation of D[O.sub.2] determinants might not be important clinically as long as the V[O.sub.2] remains above the D[O.sub.2] critical threshold. Various experts in this field have speculated that even if a D[O.sub.2] critical threshold exists in normal humans, such a relationship might not exist in the critically ill patient, or that the threshold might be at a different (higher) level in the critically ill as compared with normal humans (Fig 2). (1)
[FIGURE 2 OMITTED]
Several studies have helped elucidate these relationships further. Ronco and coworkers (11) examined the relationship between V[O.sub.2] and D[O.sub.2] in the critically ill patient. They studied nine septic and nine nonseptic patients in whom care was being withdrawn, and measured or determined V[O.sub.2], D[O.sub.2], and lactate levels. A critical threshold for Doe was identified in both groups and was approximately 3 to 5 mL/kg/min. Although one might question the ability to generalize these results because of the small patient population and selection bias (the subjects were having care withdrawn), the implication from the study is that a biphasic relationship between V[O.sub.2] and D[O.sub.2] exists in critically ill patients and the D[O.sub.2] critical threshold may be low. Friedman and associates (12) extended this work in patients with septic shock. Patients were studied both during the shock phase of their illness and during recovery. V[O.sub.2] was supply dependent in their patients without a threshold level during shock but not during recovery. Morita and colleagues (13) returned to animal models to further assess the relationship between D[O.sub.2] determinants, V[O.sub.2], and critical illness. They noted that decreased D[O.sub.2] due to a reduction in CO ("stagnant hypoxia") could be compensated for by changes in oxygen extraction only, whereas decreased D[O.sub.2] due to anemia ("anemic hypoxia") could be compensated for by increments in both oxygen extraction and CO. One postulate was that there might be different D[O.sub.2] critical thresholds depending on the mechanism of the reduction in D[O.sub.2]. D[O.sub.2] and V[O.sub.2] were therefore studied in a septic rat model (cecal ligation) and compared with control animals. Anemic hypoxia was induced by isovolemic hemodilution, and stagnant hypoxia was induced by the inflation of a balloon-tip catheter. An important finding was that the critical threshold for D[O.sub.2] was the same in septic rats regardless of the mechanism by which D[O.sub.2] was reduced but that the septic rats had a higher critical D[O.sub.2] when compared with control animals. An additional finding was that anesthesia reduced the D[O.sub.2] critical threshold.
The results of these studies suggest that a critical threshold for D[O.sub.2] exists in many seriously ill patients. While some critically ill patients may have a critical threshold value similar to normal individuals, there are subpopulations of patients in which the D[O.sub.2] critical threshold seems to be abnormal or nonexistent. The relationship of D[O.sub.2] to V[O.sub.2] may change during the course of a critical illness. Some have speculated that global measurements of V[O.sub.2] and determinations of D[O.sub.2] may be less important than regional blood flow distribution. (1,14) In patients who are critically ill with respiratory failure, the effects of changes in hemoglobin levels and CO on tissue hypoxia are difficult to predict.
CRITICAL ILLNESS: ANEMIA
Anemia is very common in the critically ill patient. (15) Although anemia may be the result of the critically ill patient's index problem (trauma or GI bleeding, for example), many patients who are hospitalized in ICUs acquire a progressive anemia over time (Fig 3). (15) This tends to be characterized as a hypoproliferative anemia with low serum iron levels, elevated or normal ferritin levels, and a low total iron binding capacity. (16) Recently reviewed by Darveau and colleagues, (17) only a handful of studies have evaluated iron metabolism in ICU patients. These studies collectively suggest that iron deficiency in ICU patients is a functional deficiency such that iron is unavailable for erythropoiesis, similar to the anemia in patients who have chronic inflammatory disease. Because iron is an essential component of bacterial growth, however, parenteral iron supplementation may be associated with an increased risk of infection and therefore cannot be universally recommended in ICU patients until further studies are performed. (17) It is thus important to first ascertain if the deficiency is a correctable one prior to initiating other therapies to treat anemia.
[FIGURE 3 OMITTED]
Patients with respiratory failure who are hospitalized in an ICU for extended periods of time while receiving mechanical ventilation frequently acquire anemia. If a patient with respiratory failure has a hemoglobin level that is low enough to reduce D[O.sub.2] to below the critical threshold, one might predict that adverse sequelae would ensue. However, one study (18) in critically ill patients suggest that a restrictive transfusion management plan is as effective in most patients as a liberal approach to transfusions. Still, most patients tend to receive transfusions at relatively liberal pretransfusion hemoglobin levels (Fig 4). (15) It becomes important, therefore, to examine specific outcomes in patients with respiratory failure in regards to transfusion therapy and anemia.
[FIGURE 4 OMITTED]
RESPIRATORY FAILURE AND TRANSFUSION THERAPY
Only a few studies are available regarding transfusion therapy and outcomes from respiratory failure or mechanical ventilation. Kahn and coworkers (19) in 1986 evaluated 15 patients with respiratory failure and anemia who were receiving mechanical ventilation and had central monitoring with a pulmonary artery catheter. Patients were administered RBC transfusions increasing their mean hemoglobin from 10.9 to 12.5 g/dL. This study (19) found that the calculated oxygen content increased, but extraction, consumption, and cardiac index did not change. The authors concluded that transfusion therapy in their patients increased calculated variables without any significant hemodynamic improvement. Habib and colleagues (20) retrospectively reviewed their experience with patients following coronary artery bypass graft (CABG) surgery. Their study group included 522 patients who were divided into an early extubation group (< 8 h of mechanical ventilation after CABG) and a late extubation group (> 8 h of mechanical ventilation after CABG). A multivariate analysis was performed that included 48 different variables to discern significant associations of these variables with prolonged mechanical ventilation, which found that receiving RBC transfusions intraoperatively or postoperatively was highly associated with prolonged mechanical ventilation. In 2001, Vamvakas and Carven (21) retrospectively reviewed records of patients undergoing CABG surgery at their institution. They identified 365 patients who were extubated within the first 24 h, and 51 patients (12.3%) who required prolonged mechanical ventilation. They noted trends toward an association between mechanical ventilation duration and the number and age of RBC units transfused.
Hebert and colleagues (22) reported outcomes of patients receiving mechanical ventilation prospectively randomized to either a "restrictive" or a "liberal" transfusion strategy. This report was a secondary subgroup analysis of a larger study (18) that involved a broad population of critically ill patients, some of whom were not receiving mechanical ventilation. The target hemoglobin level in the restrictive group was 7 to 9 g/dL, and the target in the liberal strategy group was 10 to 12 g/dL. Outcome measures that were examined included successful weaning from mechanical ventilation (defined as not receiving mechanical ventilation for 24 h), duration of mechanical ventilation, time to extubation, and ventilator-free days. Hebert and colleagues (22) found no significant differences in ventilator outcome measurements or length of ICU or hospital stay for either the entire group of patients receiving mechanical ventilation or the subgroup receiving mechanical ventilation for > 24 h.
There are few good studies focused on outcome measurements of patients receiving mechanical ventilation. Most work that has been done is retrospective. One such study by Vamvakas and Carven (23) corrected for a number of covariates related to illness severity and demonstrated a significant effect of transfusion on mechanical ventilation duration. Other studies are not controlled for comorbid illnesses and involve narrowly defined subgroups of patients. As a group, the available studies (22-25) show either no benefit from RBC transfusions or suggest that RBC transfusions are associated with poor outcomes. The poor outcomes noted may be due to associations of transfusions with other factors rather than a cause-effect relationship of transfusions to outcomes.
RESPIRATORY FAILURE AND ANEMIA
Anemia may be an associated risk factor for poor outcomes from mechanical ventilation. One of the largest studies addressing this question was published by Rady and Ryan (26) in 1999, who evaluated 11,330 patients undergoing cardiac surgery over a 3-year period. Patients were managed with a strict weaning protocol in the 24 h after surgery, and well-defined reintubation criteria were used. Six factors were identified by a multivariate analysis as being significantly associated with reintubation. Among these were a hematocrit < 34% and a transfusion requirement > 10 U of blood products. Thus, anemia and massive transfusion were associated with failed early extubation following cardiac surgery. Khamiees and coworkers (27) prospectively assessed weaning outcomes in 91 adult patients in a mixed medical-surgical ICU who successfully completed a spontaneous breathing trial. There were 100 extubations reported in the 91 patients. Sixteen patients (18%) could not be extubated, and there were two additional extubation attempts that were unsuccessful. Factors associated with extubation failure included poor cough strength, excessive secretions, and anemia. Compared to patients with higher hemoglobin levels, patients with hemoglobin levels [less than or equal to] 10 g/dL were five times as likely to experience extubation failure. Nevins and Epstein (28) retrospectively studied 166 patients with COPD who were receiving mechanical ventilation for respiratory failure. Weaning guidelines were used for all patients. Of the 166 patients, 120 survived. Risk factors associated with a poor outcome by multivariate analysis included prolonged mechanical ventilation, severe airway obstruction by spirometry, a high acute physiology and chronic health evaluation score, active malignancy, and a low hematocrit. In our ICU, we evaluated outcomes from mechanical ventilation over a 3-year period. (29) Our patient cohort included 206 patients. Adverse outcomes were considered to be death and transfer to a long-term ventilator care facility. Among 20 risk factors studied, a low hemoglobin level was found to be most significantly associated with an adverse outcome.
Anemia may be an important risk factor for poor outcome from respiratory failure and mechanical ventilation, but there are very limited scientific data available. Studies that have been reported are mainly small retrospective analyses that are not controlled for comorbid illness.
ERYTHROPOIETIN AND MECHANICAL VENTILATION OUTCOMES
Corwin and colleagues (30) studied outcomes in critically ill patients who were randomized to receive either erythropoietin or placebo. The study drug was administered once weekly by subcutaneous injection on ICU day 3 and then weekly thereafter. Administration of erythropoietin increased hemoglobin and reduced the need for allogeneic RBC transfusions when compared with control subjects. Median ventilator-free days, reventilation rates, and new-onset mechanical ventilation did not differ between the study and the control populations. This study enrolled a heterogeneous group of critically ill patients, and the mechanical ventilator outcomes were secondary end points. Further study would be needed to fully assess the impact of erythropoietin administration on mechanical ventilation outcomes in the critically ill. If erythropoietin therapy is to be considered in critically ill patients as an .alternative to transfusions, it should be kept in mind that consumption of iron occurs during heme synthesis. Although controversial, as discussed above, iron supplementation might be needed to keep pace with the demand imposed by the stimulation of erythropoiesis. (31)
CONCLUSIONS
Respiratory failure requiring support by mechanical ventilation is a common problem encountered in caring for critically ill patients. The optimization of D[O.sub.2] is a logical strategy in such patients. To the extent that this approach is important, anemia and RBC transfusions might be expected to have an important effect on outcomes in critically ill patients. Because of the biphasic relationship of D[O.sub.2] to V[O.sub.2] improvements in D[O.sub.2] may not affect V[O.sub.2] unless the D[O.sub.2] is below a critical threshold level. The critical threshold for Dos may be altered or nonexistent in some critically ill patients.
Some data exist suggesting that anemia is associated with adverse outcomes in patients receiving mechanical ventilation, including poorer survival, increased duration of mechanical ventilation, and increased reintubation rates. However, few studies in anemic, critically ill patients are controlled for severity or nature of illness, and most studies are small and retrospective or are secondary analyses. There is little evidence to suggest that BBC transfusions in anemic patients receiving mechanical ventilation improve outcome. In fact, some data exist supporting an association between RBC transfusions and poorer ventilation and ICU outcomes. Most of the studies involving transfusions are also not controlled for severity of illness.
The reasons why RBC transfusions are ineffective in changing outcomes in the critically ill are not dear. One possibility is that the critical threshold for D[O.sub.2] is too low in most critically ill patients for transfusions to lead to important increases in V[O.sub.2] and utilization. (1,6,10,12,13) Stored blood with a long shelf life may be a relatively ineffective medium for oxygen transport because of decreased (2,3)-diphosphoglycerate or decreased RBC deformability. (32)
Complications specific to the respiratory system may also occur with RBC transfusions, limiting the realization of positive outcomes. Such complications might include volume overload, transfusion-related lung injury, or increased rates of nosocomial pneumonia. (33,34)
Most studies involving patients with respiratory failure, anemia, and transfusions are small and retrospective, or use these end points as secondary outcomes. More research should be done in specific populations of patients receiving mechanical ventilation in regards to anemia, RBC transfusions, and specific ventilator outcomes. Technology to improve the safety and efficacy of RBC transfusions in critically ill patients should be explored. New treatments for anemia in the critically ill should be developed. Therapies such as erythropoietin administration offer promise, but these treatments need to be evaluated in reference to mechanical ventilation outcomes.
REFERENCES
(1) Leach RM, Treacher DF. The pulmonary physician and critical care: 6. Oxygen transport: the relation between oxygen delivery and consumption. Thorax 1992; 47:971-978
(2) Consensus Conference in Intensive Care Medicine. Third European Consensus Conference in Intensive Care Medicine. Tissue hypoxia: how to detect, how to correct, how to prevent. Societe de Reanimation de Langue Francaise. The American Thoracic Society. European Society of Intensive Care Medicine. Am J Respir Crit Care Med 1996; 154:1573-1578
(3) American Thoracic Society. Tissue hypoxia: how to detect, how to correct, how to prevent? J Crit Care 1997; 12:39-47
(4) Leach RM, Treacher DF. Oxygen transport: 2. Tissue hypoxia. BMJ 1998; 317:1370-1373
(5) Treacher DF, Leach RM. Oxygen transport: 1. Basic principles. BMJ 1998; 317:1302-1306
(6) Schliehtig R. Oxygen delivery and consumption in critical illness. In: Civetta JM, Taylor RW, Kirby RR, eds. Critical care. Philadelphia, PA: Lippincott-Raven, 1997; 337-342
(7) Reeves JT, Groves BM, Sutton JR, et al. Operation Everest II: preservation of cardiac function at extreme altitude. J Appl Physiol 1987; 63:531-539
(8) Weiskopf RB, Viele MK, Feiner J, et al. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA 1998; 279:217-221
(9) Hsia CC. Respiratory function of hemoglobin. N Engl J Med 1998; 338:239-247
(10) Cain SM. Oxygen delivery and uptake in clogs during anemic and hypoxic hypoxia. J Appl Physiol 1977; 42:228-234
(11) Ronco JJ, Fenwick JC, Tweeddale MG, et al. Identification of the critical oxygen delivery for anaerobic metabolism in critically ill septic and nonseptic humans. JAMA 1993; 270: 1724-1730
(12) Friedman G, De Backer D, Shahla M, et al. Oxygen supply dependency can characterize septic shock. Intensive Care Med 1998; 24:118-123
(13) Morita Y, Chin-Yee I, Yu P, et al. Critical oxygen delivery in conscious septic rats under stagnant or anemic hypoxia. Am J Respir Crit Care Med 2003; 167:868-872
(14) Doglio GR, Pusajo JF, Egurrola MA, et al. Gastric mucosal pH as a prognostic index of mortality/in clinically ill patients. Crit Care Med 1991; 19:1037-1040
(15) 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
(16) 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
(17) Darveau M, Denault AY, Blais N, et al. Bench-to-bedside review: iron metabolism in critically ill patients. Crit Care 2004; 8:356-62
(18) 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
(19) Kahn RC, Zaroulis C, Goetz W, et al. Hemodynammic oxygen transport and 2,3-diphospboglycerate changes after transfusion of patients in acute respiratory failure. Intensive Care Med 1986; 12:22-25
(20) Habib RH, Zacharias A, Engoren M. Determinants of prolonged mechanical ventilation after coronary artery bypass grafting. Ann Thorac Surg 1996; 62:1164-1171
(21) Vamvakas EC, Carven JH. Allogeneic blood transfusion and postoperative duration of mechanical ventilation. Transfusion 2001; 41:885-892
(22) Hebert PC, Blajchman MA, Cook DJ, et al. Do blood transfusions improve outcomes related to mechanical ventilation? Chest 2001; 119:1850-1857
(23) Vamvakas EC, Carven JH. Allogeneic blood transfusion and postoperative duration of mechanical ventilation: effects of red cell supernatant, platelet supernatant, plasma components and total transfussed fluid. Vox Sang 2002; 82:141-149
(24) Levy EC MacIntyre N, Abraham E. Anemia, blood transfusion (Tx) and ICU outcomes in mechanically ventilated (MV) vs non-MV (nMV) patients [abstract]. Am J Respir Crit Care Med 2003; 167:A22
(25) Webber GV, Sukumaran M. Blood transfusion requirements in mechanically ventilated ICU patients [abstract]. Am J Respir Crit Care Med 2002; 165:A792
(26) Rady MY, Ryan T. Perioperative predictors of extubation failure and the effect on clinical outcome after cardiac surgery. Crit Care Med 1999; 27:340-347
(27) Khamiees M, Raju P, DeGirolamo A, et al. Predictors of extubation outcome in patients who have successfully completed a spontaneous breathing trial. Chest 2001; 120:1262-1270
(28) Nevins ML, Epstein SK. Predictors of outcome for patients with COPD requiring invasive mechanical ventilation. Chest 2001; 119:1840-1849
(29) Ouellette DR, Quinn MW, Emmons EF, et al. Decreased hemoglobin associated with impaired weaning from mechanical ventilation [abstract]. Am J Flespir Crit Care Med 2000; 161:A560
(30) 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
(31) Lapointe M. Iron supplementation in the intensive care unit: when, how much, and by what route? Crit Care 2004; 8(suppl):S37-S41
(32) Marik PE, Sibbald WJ. Effect of stored-blood transition on oxygen delivery in patients with sepsis. JAMA 1993; 269: 3024-3029
(33) Goodnough LT, Brecher ME, Kanter MH, et al. Transition medicine: first of two parts; blood transition. N Engl J Med 1999; ,340:438-447
(34) Shorr AF, Duh M-S, Kelly KM, et al. Red blood cell transition and ventilator-associated pneumonia: a potential link? Crit Care Med 2004; 32:666-674
* From the Department of Pulmonary and Critical Care Medicine, Brooke Army Medical Center, Fort Sam Houston, TX. This publication was supported by an educational grant from Ortho Biotech Products, L.P.
The following authors have disclosed financial relationships with a commercial party. Grant information and company names appear as provided by the author: Daniel R. Ouellette, MD, FCCP: Ortho Biotech--Speaker bureau.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/mise/reprints.shtml).
Correspondence to: Daniel R. Ouellette, MD, FCCP, Fellowship Training Director for Pulmonary and Critical Care Medicine, Brooke Army Medical Center, 3851 Roger Brooke Dr, Fort Sam Houston, TX 78234; e-mail: DanielROuellette@aol.com
COPYRIGHT 2005 American College of Chest Physicians
COPYRIGHT 2005 Gale Group