Blood Conservation in Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome

ADULT CARDIAC

Blood Conservation in Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome Cara L. Agerstrand, MD, Kristin M. Burkart, MD, MS, Darryl C. Abrams, MD, Matthew D. Bacchetta, MD, MBA,* and Daniel Brodie, MD* Division of Pulmonary, Allergy, and Critical Care Medicine, Department of Medicine, and Division of Cardiothoracic Surgery, Department of Surgery, Columbia University College of Physicians and Surgeons/NewYork-Presbyterian Hospital, New York, New York

Background. Extracorporeal membrane oxygenation support (ECMO) typically requires multiple blood transfusions and is associated with frequent bleeding complications. Blood transfusions are known to increase morbidity and mortality in critically ill patients, which may extend to patients receiving ECMO. Aiming to reduce transfusion requirements, we implemented a blood conservation protocol in adults with severe acute respiratory distress syndrome (ARDS) receiving ECMO. Methods. This was a retrospective study of adults receiving ECMO for ARDS after initiation of a blood conservation protocol that included a transfusion trigger of hemoglobin of less than 7.0 g/dL, use of low-dose anticoagulation targeting an activated partial thromboplastin time of 40 to 60 seconds, and autotransfusion of circuit blood during decannulation. The primary objective was to evaluate transfusion requirements during ECMO support. Clinical outcomes included survival, neurologic function, renal function, bleeding, and thrombotic complications.

Results. The analysis included 38 patients; of these, 24 (63.2%) received a transfusion while receiving ECMO. Median hemoglobin was 8.29 g/dL. A median of 1.0 units (range, 250 to 300 mL) was transfused during ECMO support over a median duration of 9.0 days, equivalent to 0.11 U/d (range, 27.5 to 33.3 mL/d). The median activated partial thromboplastin time was 46.5 seconds. Bleeding occurred in 10 patients (26.3%); severe bleeding occurred in 2 patients (5.3%). Twenty-eight patients (73.7%) survived to hospital discharge. Conclusions. Implementation of a blood conservation protocol in adults receiving ECMO for ARDS resulted in lower transfusion requirements and bleeding complications than previously reported in the literature and was associated with comparable survival and organ recovery.

A

threshold remains a source of controversy. Compared with conservative approaches to blood transfusion that accept a moderate degree of anemia and a hemoglobin as low as 7.0 g/dL, liberal strategies that transfuse to normal or nearly normal hemoglobin levels increase morbidity and mortality in critically ill patients [11–14]; this may extend to patients receiving ECMO. Traditional management of ECMO patients also includes high levels of anticoagulation [4] and is associated with frequent bleeding complications [2, 4]. To reduce transfusions, we implemented a three-part blood conservation protocol for patients receiving ECMO for ARDS that included use of a restrictive transfusion strategy [12]. We accept a hemoglobin level as low as 7.0 g/dL, use low-dose anticoagulation, and autotransfuse the blood within the circuit at the time of ECMO decannulation, thereby preserving RBC mass and minimizing blood loss to the circuit. We examined

dvances in extracorporeal membrane oxygenation (ECMO) technology have made it safer and easier to use [1]. Interest in ECMO surged during the 2009 influenza A (H1N1) pandemic in adult patients with severe acute respiratory distress syndrome (ARDS) [2] and after the publication of a randomized controlled trial of ECMO for adults with acute respiratory failure [3]. The optimal medical management of patients receiving ECMO, including the transfusion threshold and the degree of anticoagulation, has not been established, however. The Extracorporeal Life Support Organization recommends that patients receive a transfusion to a normal hemoglobin of 12 to 14 g/dL [4], which frequently requires transfusion of 2 to 3 units of packed red blood cells (pRBCs) daily [5–9], although requirements of up to 6 U/d have been reported [10]. Many ECMO centers follow this approach, although the optimal transfusion

(Ann Thorac Surg 2015;99:590–6) Ó 2015 by The Society of Thoracic Surgeons

Accepted for publication Aug 25, 2014. *Drs Brodie and Bacchetta are co-senior authors. Address correspondence to Dr Brodie, 622 W 168th St, PH 8-101, New York, NY 10032; e-mail: [email protected].

Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier

Dr Brodie discloses financial relationships with Maquet Cardiovascular and ALung Technologies.

0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.08.039

Ann Thorac Surg 2015;99:590–6

Abbreviations and Acronyms

our transfusion requirements, bleeding complications, and clinical outcomes after the implementation of this approach.

Material and Methods This study, which was approved by the Columbia University Investigational Review Board and performed in accordance with accepted ethical standards, included adults who received ECMO for ARDS in the medical intensive care unit at New York-Presbyterian Hospital/ Columbia University College of Physicians and Surgeons between January 1, 2010, and December 31, 2012, the initial 3-year period after the formal adoption of our blood conservation protocol. We hypothesized that this approach would result in fewer pRBC transfusions and bleeding complications than previously reported. The primary objective was to quantify the units of pRBCs transfused during ECMO support. Secondary objectives were to characterize bleeding and thrombotic complications, survival to intensive care unit and hospital discharge, and neurologic and renal function after recovery. ARDS was defined according to the Berlin criteria [15]. The decision to initiate ECMO support was at the discretion of our center’s ECMO team, which included a pulmonary intensivist and thoracic surgeon. Patients considered for ECMO met criteria for severe ARDS with any of the following: a partial pressure of arterial oxygen– to–fraction of inspired oxygen ratio of less than 80 mm Hg, uncompensated hypercapnia with a pH of less than 7.15, or excessively high plateau pressures exceeding 35 to 45 cm H2O, despite optimized ventilator management [1]. Most patients were cannulated by our center’s mobile ECMO transport team and transferred to our institution on ECMO; therefore, use of other rescue therapies varied by availability at the patient’s originating hospital. Relative contraindications to ECMO included advanced age, prolonged high-pressure ventilation, or any condition limiting the likelihood of benefit from ECMO, such as irreversible neurologic injury or metastatic cancer [1]. Shock, renal failure, acute liver injury, known bleeding

591

diatheses, trauma (including traumatic brain injury), or refusal to accept blood products were not considered contraindications to ECMO. Patients were managed with a three-part blood conservation protocol that included a RBC transfusion trigger of hemoglobin of less than 7.0 g/dL, an activated partial thromboplastin time (aPTT) goal between 40 and 60 seconds, and autotransfusion of circuit blood during decannulation. Anticoagulation was administered by continuous infusion and titrated according to a cliniciandriven protocol. Patients diagnosed with a deep venous thrombosis were managed with therapeutic levels of intravenous anticoagulation. Aspirin was not routinely used. Although not a formal component of our protocol, our center takes a parsimonious approach to phlebotomy. The frequency of laboratory monitoring is at the clinician’s discretion and occurs at least daily; gases are not routinely measured before and after oxygenator initiation. Throughout the study period, ECMO circuit components included a centrifugal pump, polymethylpentene oxygenator, heat exchanger, and monitors of pressure before and after oxygenator initiation. Demographic, clinical, and laboratory data were obtained from our institution’s electronic medical record. The median hemoglobin was obtained during the period the patient received ECMO. Transfusion requirements were measured as the number of units and milliliters of pRBCs transfused. At our institution, each unit of blood contains 250 to 300 mL of pRBCs. Data are presented as median values with interquartile range (IQR). The glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease Equation [16]. Normal GFR was reported as exceeding 60 mL/min/1.73 m2 [17]. Data were analyzed with SAS 9.2 software (SAS Institute Inc, Cary, NC) and Excel 14.2.2 software (Microsoft Corp, Redmond, WA).

Results During the study period, 38 patients received ECMO for ARDS. Demographic data and pre-ECMO characteristics are listed in Table 1. Initial ECMO support was venovenous in 34 patients (89.4%), venoarterial in 2 (5.3%), and venoarterial venous in 2 (5.3%; Table 2). Median ECMO blood flow before weaning was 4.1 L/min (IQR, 3.8 to 4.7 L/ min). The median ratio of ECMO blood flow to predicted cardiac output was 0.87 (IQR, 0.8 to 1). Median sweep gas flow was 3.4 L/min (IQR, 2.3 to 5.2 L/min), fraction of delivered oxygen was 1.0, and the preoxygenator saturation was 74.5% (IQR, 68.1% to 81%). The median duration of ECMO support was 9 days (IQR, 7 to 11.5 days). The median hemoglobin during ECMO support was 8.2 g/dL (IQR 7.9 to 8.7 g/dL), with values of 9.1 g/dL at cannulation and 8.1 g/dL at decannulation (Table 3). At the time of cannulation, the study population had a platelet count of 163,000/mL (IQR 128,500 to 219,000/mL), a prothrombin time of 17.5 seconds (IQR, 15.8 to 18.4 seconds), international normalized ratio of 1.4 (IQR, 1.2 to 1.5),

ADULT CARDIAC

APACHE II = Acute Physiology and Chronic Health Evaluation II aPTT = activated partial thromboplastin time ARDS = acute respiratory distress syndrome ECMO = extracorporeal membrane oxygenation = fraction of delivered oxygen FDO2 = fraction of inspired oxygen FIO2 GFR = glomerular filtration rate IQR = interquartile range PaO2 = partial pressure of arterial oxygen PEEP = positive end-expiratory pressure pRBCs = packed red blood cells

AGERSTRAND ET AL BLOOD CONSERVATION IN ECMO FOR ARDS

592

AGERSTRAND ET AL BLOOD CONSERVATION IN ECMO FOR ARDS

Table 1. Baseline Demographics

ADULT CARDIAC

Variables Age, y Male sex Body mass index, kg/m2 APACHE II score Comorbidities Hypertension Organ transplantation Trauma Chronic kidney disease Diabetes mellitus Pregnancy/postpartum Traumatic brain injury Congenital heart disease Pneumonectomy Other immunosuppression Collagen vascular disease Interstitial lung disease Etiology of ARDS Pneumonia Abdominal sepsis Interstitial lung disease with ARDS Eosinophilic pneumonia Acute cellular rejection Transfusion-related ARDS Respiratory parameters before ECMO PaO2/FIO2 PEEP, cm H2O Salvage therapies before ECMO Paralysis Inhaled nitric oxide Inhaled epoprostenol Prone positioning High-frequency oscillatory ventilation

Median (IQR) of No. (%) (N ¼ 38) 33 24 28.0 23 8 7 6 6 3 3 2 2 2 1 1 1

(24–53) (63.2) (22.7–32.0) (18–27) (21.1) (18.4) (15.8) (15.8) (7.9) (7.9) (5.3) (5.3) (5.3) (2.6) (2.6) (2.6)

32 (84.2) 2 (5.3) 2 (5.3) 1 (2.6) 1 (2.6) 1 (2.6)

53 (45–63) 15 (12–15) 24 9 2 3 1

(63.2) (23.7) (5.3) (7.9) (2.6)

APACHE II ¼ Acute Physiology and Chronic Health Evaluation II; ARDS ¼ acute respiratory distress syndrome; ECMO ¼ extracorporeal membrane oxygenation; FIO2 fraction of inspired oxygen; IQR ¼ interquartile range; PaO2 ¼ partial pressure of arterial oxygen; PEEP ¼ positive end-expiratory pressure.

aspartate aminotransferase of 45 U/L (IQR, 33.8 to 94.3 U/L), and total bilirubin of 0.9 mg/dL (IQR, 0.5 to 1.3 mg/dL). Blood transfusions were administered to 24 patients (63.2%). The median quantity blood transfused was 1.0 unit (IQR, 0 to 3 units), equivalent to 250 mL (IQR, 0 to 750 mL) to 300 mL (IQR, 0 to 900 mL) of pRBCs, over the entire duration of ECMO support, or 0.11 units or 27.8 mL (IQR 0 to 83.3 mL) to 33.3 mL (IQR, 0 to 100 mL) daily (Table 3). Very low transfusion requirements were observed throughout the study period. There was no relationship between transfusion requirements and year of study entry. The average age of the blood transfused during a chronologic subset of these patients was 28.4 days.

Ann Thorac Surg 2015;99:590–6

Table 2. Details of Extracorporeal Membrane Oxygenation Support Variables

No. (%) or Median (IQR) (N ¼ 38)

Initial ECMO configuration Venovenous Venoarterial Venoarterial-venous Intubation before ECMO initiation, d ECMO duration, d ECMO variables Blood flow, L/min Sweep gas flow, L/min FDO2, % Preoxygenator saturation, %

34 2 2 3 9

(89.4) (5.3) (5.3) (1.3–6) (7–11.5)

4.1 (3.8–4.7) 3.4 (2.3–5.2) 100 74.5 (68.1–81)

ECMO ¼ extracorporeal membrane oxygenation; delivered oxygen; IQR ¼ interquartile range.

FDO2 ¼ fraction of

The median aPTT was 46.5 seconds (IQR, 41.8 to 50.8 seconds) and was 45.5 seconds in the nontransfused group and 46.5 seconds in the transfused group (p ¼ 0.83). Despite the low target aPTT, circuit or oxygenator thrombosis requiring replacement was rare, occurring in only 1 patient who had a hypercoagulable state from underlying rheumatologic disease. Clinically apparent bleeding occurred in 10 patients (26.3%), most of whom had self-limited oozing at cannula or other surgical sites (Table 4). Two severe bleeding complications occurred: hemoptysis from alveolar hemorrhage and severe cannula-site bleeding, both of which prompted early ECMO decannulation. There was no gastrointestinal, retroperitoneal, intraabdominal, or intracranial bleeding. The remainder of the transfusions occurred secondary to a slow decline in hemoglobin attributed to critical illness and phlebotomy. Clinically significant hemolysis, as determined by declining hemoglobin in the setting of suggestive laboratory data (elevated indirect hyperbilirubinemia, elevated lactate dehydrogenase, or reduced haptoglobin) did not occur. Table 3. Hematologic Variables During Extracorporeal Membrane Oxygenation Support Variables Hemoglobin, g/dL pRBCs transfused Total per patient Units Volume, mL Per day of ECMO support Units Volume, mL aPTT, s

Median (IQR) (N ¼ 38) 8.2 (7.9–8.7)

1 (0–3) 250 (0–750)–300 (0–900) 0.11 (0–0.33) 27.8 (0–83.3)–33.3 (0–100) 46.5 (40.8–50.8)

aPTT ¼ activated partial thromboplastin time; ECMO ¼ extracorporeal membrane oxygenation; IQR ¼ interquartile range; pRBC ¼ packed red blood cells.

Ann Thorac Surg 2015;99:590–6

AGERSTRAND ET AL BLOOD CONSERVATION IN ECMO FOR ARDS

Table 4. Clinical Outcomes Outcomes

a

28 28 12 3 10 4 4 2 2 1 8 4 4

(73.7) (73.7) (31.6) (7.9) (26.3) (10.5) (10.5) (5.3) (5.3) (2.6) (23.5)a (11.8)a (11.8)a

Values are for 34 patients screened for deep venous thrombosis.

Screening vascular ultrasound scans were performed in 34 patients (89%). Screening in survivors occurred after ECMO decannulation; ultrasound scans were performed in most nonsurvivors during their ECMO runs, although at variable times. Cannula-related deep venous thromboses were diagnosed in 8 patients (23.5% of those screened), including 4 nonocclusive and 4 occlusive thromboses at ECMO cannulation sites (Table 4). Six of the 8 patients (75%) diagnosed with deep venous thrombosis survived. Twenty-eight patients (73.7%) survived to intensive care unit and hospital discharge (Table 4). Twelve patients (31.6%) required renal replacement therapy due to acute renal failure that developed before or at the time of ECMO initiation. Among survivors, 21 (75%) had a normal GFR (>60 mL/min/1.73 m2) and 4 (14.3%) were discharged with impaired but improving GFR. Three survivors (10.7%) required hemodialysis at discharge; of whom, 2 had known chronic kidney disease, with baseline creatinine between 2 and 3 mg/dL. All survivors had a Modified Glasgow Coma Scale score of 15.

Comment The principal purpose of blood transfusions is to increase oxygen delivery to tissues; however, there is conflicting evidence about the ability of transfused blood to improve oxygen delivery, despite the associated increase in hemoglobin [18–20]. A recent report of a series of patients who received venovenous ECMO for ARDS noted that transfusions increased oxygen content, oxygen delivery, and allowed for a reduction in ECMO blood flow rate [21]. However, any potential benefit of transfusion must be weighed against the risk, because multiple clinical studies demonstrate an association between transfusions and worsened outcomes, including increased death, transfusion-related acute lung injury, worsened ARDS, volume overload, and poor wound healing [11–14, 22–24].

The number of transfusions, not baseline hemoglobin, is associated with higher intensive care unit and hospital death [11, 13]. Transfusion-associated morbidity extends to patients receiving even 1 to 2 units of blood and increases in a dose-dependent manner [14]. The multicenter, randomized Transfusion Requirements in Critical Care trial demonstrated a significant decrease in death with a restrictive transfusion strategy (hemoglobin trigger of 7.0 g/dL) compared with a liberal strategy (hemoglobin trigger of 10.0 g/dL) among patients aged younger than 55 years and an Acute Physiology and Chronic Health Evaluation II score of less than 20 [12]. On the basis of current evidence, the 2012 guidelines of the American Association of Blood Banks recommend use of a restrictive transfusion strategy (hemoglobin 7.0 to 8.0 g/dL) for hemodynamically stable, hospitalized patients [25]. Mechanisms of transfusion-related organ dysfunction are poorly understood. Blood storage age may be a factor [19, 26–29] and may be mediated by host-pathogen interactions [30]. Pooled RBCs undergo multiple biochemical and structural changes that impair functioning once transfused, including decreased microcirculatory flow and increased endothelial adherence, cytokine release, and neutrophil activation, which may interfere with oxygen delivery [31–33]. Furthermore, RBC 2,3 diphosphoglycerate decreases with time and results in increased oxygen-binding affinity to hemoglobin and reduced tissue uptake [34]. Free hemoglobin is increased in stored RBCs and may lead to vasoconstriction and intravascular thrombosis [32, 34]. Blood transfused to adults could have a storage time of 3 weeks or longer [11]. Our center’s three-part blood conservation protocol for adults receiving ECMO for ARDS is rooted in the evidence that supports a restrictive transfusion strategy in other critically ill patients. Of note, our patient population was relatively young, without significant medical comorbidities such as coronary artery disease; the transfusion threshold in patients with known coronary artery disease is less well established [35, 36]. Our approach differs substantially from traditional practice in which patients receiving ECMO are given transfusions to normal or nearly normal hemoglobin, commonly receiving multiple transfusions daily [5–9]. Our median pRBC transfusion of 1 unit, equivalent to 250 to 300 mL, over the entire duration of ECMO support, or 0.11 units (27.5 to 33.3 mL) daily, is the lowest reported in the literature and less than 10% of historical transfusion rates [5–9]. Bleeding is reduced through use of a low-dose anticoagulation protocol targeting an aPTT between 40 and 60 seconds. In patients with known bleeding, we target an aPTT at the lower end of this range. Our low rates of severe bleeding complications (5%) and bleeding of any kind (26%) compare favorably with the bleeding rates of 27% to 54% reported in other recent series of ECMO patients, in which more episodes of severe bleeding were also reported [37, 38]. The reduced risk of bleeding associated with anticoagulation targets must be balanced against the potentially increased risk of thrombosis or oxygenator malfunction. The optimal level of anticoagulation required to prevent thrombosis in the newer

ADULT CARDIAC

Survival to Intensive care unit discharge Hospital discharge Renal replacement therapy Hemodialysis at hospital discharge Bleeding—all sites Cannula site Pulmonary Tracheostomy site Nasal or oropharyngeal Hematuria Cannula-related deep venous thrombosis Occlusive Partially occlusive

No. (%) (N ¼ 38)

593

ADULT CARDIAC

594

AGERSTRAND ET AL BLOOD CONSERVATION IN ECMO FOR ARDS

ECMO circuits is not yet known. In our experience, however, circuit thrombosis and oxygenator malfunction at this level of anticoagulation are rare. We further preserve RBC mass at the time of ECMO decannulation when we return approximately 80% of the ECMO circuit blood back to the patient. The anticipated return of circuit blood allows us to avoid transfusions near the time of decannulation and further reduces the need for pRBC transfusion. In addition, our center uses a simplified ECMO circuit consisting of only the key component parts. Compared with older and more complex ECMO circuits, clinically significant hemolysis, defined by decreasing hemoglobin in the setting of consistent laboratory values, has not definitively occurred in our cohort. The optimal hemoglobin level for patients with severe ARDS is not well studied and is undefined for patients with severe ARDS receiving ECMO. However, our high survival rates with a hemoglobin transfusion trigger of 7.0 g/dL suggest that a conservative transfusion strategy may be applicable to this population. To optimize oxygen delivery in the setting of relatively lower hemoglobin, we target an oxygen saturation of 92% or higher and a high ECMO blood flow rate relative to a patient’s predicted cardiac output. The higher the proportion of ECMO blood flow to cardiac output, the greater the relative contribution of ECMO to systemic oxygenation. The median ratio of ECMO blood flow to predicted cardiac output in our cohort was 0.87 (IQR, 0.8 to 1). Because actual cardiac output may be higher than predicted cardiac output in a critically ill population, this ratio may overestimate the contribution of ECMO to a patient’s systemic oxygen delivery. The median preoxygenator saturation of 74.5% suggests that oxygen delivery was adequate despite the restrictive transfusion protocol and median hemoglobin of 8.2 g/dL. Neurologic and renal function at hospital discharge may be surrogates of overall organ function and evidence of adequate oxygen delivery; however, these also depend on other variables such as hemodynamics and baseline function. Three studies using modern ECMO technology in patients with severe ARDS reported lower rates of pRBC transfusion than prior studies [37–39]. However, their transfusion rates were considerably higher than our cohort: The Australia and New Zealand ECMO Investigators reported a median of 1,880 mL (IQR, 904 to 3,750 mL) of blood transfused over a median of 10 days (IQR, 7 to 15 days) of ECMO support, equivalent to 188 mL/d (0.75 U/d) [38]. A group of Italian ECMO centers reported a median of 1,500 mL (IQR, 400 to 2,990 mL) of pRBCs transfused over the entire duration of ECMO treatment, or 155 mL (IQR, 41.3 to 309 mL) daily (0.62 units daily) [37]. Survival in influenza A (H1N1)-associated ARDS, as in these two studies, tends to be higher than in the general ARDS population, such as ours [2]. Nevertheless, our survival rate of 73.6% is comparable to the 75% reported in the Australia and New Zealand ECMO study and 68% reported in the Italian study, with otherwise similar patient

Ann Thorac Surg 2015;99:590–6

populations [37, 38, 40]. Larger, prospective trials are needed to determine the optimal hemoglobin and anticoagulation targets in patients receiving ECMO. In conclusion, the potentially harmful consequences of transfusions and the evidence supporting a conservative approach to transfusions in critically ill patients may extend to patients receiving ECMO. Our experience suggests that a blood conservation protocol, including a restrictive approach to transfusions, is effective at reducing transfusion requirements. A low-dose anticoagulation protocol appears to be safe, effective, and may reduce bleeding complications. Although this study is limited by its retrospective, single-center design and small sample size, our favorable clinical outcomes suggest a restrictive transfusion strategy is a reasonable approach in patients receiving ECMO for ARDS.

References 1. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J Med 2011;365:1905–14. 2. Paden ML, Conrad SA, Rycus PT, Thiagaragan RR. Extracorporeal Life Support Organization Registry Report 2012. ASAIO J 2013;59:202–10. 3. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 2009;374:1351–63. 4. Annich GM, Lynch WR, MacLaren G, Wilson JM, Bartlett RH (eds). ECMO: extracorporeal cardiopulmonary support in critical care. 4th edition. Ann Arbor: Extracorporeal Life Support Organization, 2012. 5. Lewandowski K, Rossaint R, Pappert D, et al. High survival rate in 122 ARDS patients managed according to a clinical algorithm including extracorporeal membrane oxygenation. Intensive Care Med 1997;23:819–35. 6. Mols G, Loop T, Geiger K, Farthmann E, Benzing A. Extracorporeal membrane oxygenation: a ten-year experience. Am J Surg 2000;180:144–54. 7. Nehra D, Goldstein AM, Doody DP, Ryan DP, Chang Y, Masiakos PT. Extracorporeal membrane oxygenation for nonneonatal acute respiratory failure: the Massachusetts General Hospital experience from 1990 to 2008. Arch Surg 2009;144:427–32. 8. Norfolk SG, Hollingsworth CL, Wolfe CR, et al. Rescue therapy in adult and pediatric patients with pH1N1 influenza infection: a tertiary center intensive care unit experience from April to October 2009. Crit Care Med 2010;38:2103–7. 9. Butch SH, Knafl P, Oberman HA, Bartlett RH. Blood utilization in adult patients undergoing extracorporeal membrane oxygenated therapy. Transfusion 1996;36:61–3. 10. Gattinoni L, Pesenti A, Mascheroni D, et al. Low-frequency positive-pressure ventilation with extracorporeal C02 removal in severe acute respiratory failure. JAMA 1986;256: 881–6. 11. 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. 12. H ebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 1999;340:409–17. 13. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA 2002;288:1499–507. 14. Ferraris VA, Davenport DL, Saha SP, Bernard A, Austin PC, Zwischenberger JB. Intraoperative transfusion of small amounts of blood heralds worse postoperative outcome in

Ann Thorac Surg 2015;99:590–6

15.

17.

18.

19.

20. 21.

22. 23. 24. 25. 26. 27.

28.

29. 30.

31. 32. 33.

34. 35. 36. 37.

38. 39. 40.

595

coagulation status: a randomized clinical trial. Am J Respir Crit Care Med 2012;185:842–50. Spinella PC, Carroll CL, Staff I, et al. Duration of red blood cell storage is associated with increased incidence of deep vein thrombosis and in hospital mortality in patients with traumatic injuries. Crit Care 2009;13:R151–61. Pettila V, Westbrook AJ, Nichol AD, et al. Age of red blood cells and mortality in the critically ill. Crit Care 2011;15: R116–23. Tung JP, Fraser JF, Nataatmadja M, et al. Age of blood and recipient factors determine the severity of transfusionrelated acute lung injury (TRALI). Crit Care 2012;16: R19–32. Lelubre C, Vincent JL. Red blood cell transfusion in the critically ill patient. Ann Intensive Care 2011;1:43–51. Tinmouth A, Fergusson D, Yee IC, Hebert PC. Clinical consequences of red cell storage in the critically ill. Transfusion 2006;46:2014–27. Zhu H, Zennadi R, Xu BX, et al. Impaired adenosine-50 triphosphate release from red blood cells promotes their adhesion to endothelial cells: a mechanism of hypoxemia after transfusion. Crit care Med 2011;39:2478–86. Bennett-Guerrero E, Veldman TH, Doctor A, et al. Evolution of adverse changes in stored RBCs. Proc Natl Acad Sci U S A 2007;104:1706–68. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery. JAMA 2010;304: 1559–67. H ebert PC, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001;29:227–34. Patroniti N, Zangrillo A, Pappalardo F, et al. The Italian ECMO network experience during the 2009 influenza A (H1N1) pandemic: preparation for severe respiratory emergency outbreaks. Intensive Care Med 2011;37: 1447–57. Davies A, Jones D, Bailey M, et al. Extracorporeal membrane oxygenation for 2009 influenza A (H1N1) acute respiratory distress syndrome. JAMA 2009;302:1888–95. Muller T, Philipp A, Luchner A, et al. A new miniaturized system for extracorporeal membrane oxygenation in adult respiratory failure. Crit Care 2009;13:R205–14. Davies A, Jones D, Gattas D. Extracorporeal membrane oxygenation for ARDS due to 2009 influenza A (H1N1)— Reply. JAMA 2010;303:942.

INVITED COMMENTARY This study by Agerstrand and colleagues [1] addresses an important problem: blood product management during extracorporeal membrane oxygenation (ECMO) for acute respiratory distress syndrome (ARDS). Historical practice has been to transfuse patients receiving ECMO to nearly normal hemoglobin levels, but accumulating data in a variety of critical illnesses suggest that lower hemoglobin targets (eg, 7 g/dL) are associated with equivalent or better outcomes. Blood products are an expensive and limited resource, and in ARDS it makes sense to avoid unnecessary transfusions to minimize volume administration and avoid transfusion-related lung injury. A balance must be struck between the benefits of increased oxygen delivery at higher hemoglobin concentrations and the deleterious effects of blood transfusion. This study informs debate on that balance.

Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier

To reduce transfusions of packed red blood cells (PRBCs), Agerstrand and colleagues used a 3-part blood conservation protocol: (1) a PRBC transfusion trigger of hemoglobin less than 7 g/dL, (2) low-dose anticoagulation (target activated partial thromboplastin time 40–60 seconds), and (3) autotransfusion of circuit blood during ECMO decannulation. Most patients (34 of 38 [89.4%]) were supported with venovenous ECMO. The majority of patients (63.2%) received PRBCs during the ECMO run, but the median quantity of transfusion was only 1 unit (range, 0–3 units) over the duration of ECMO support (median 9 days). Overall, 28 of the 38 patients (73.7%) survived to hospital discharge. The use of ECMO for ARDS is increasing, and best practices have yet to be defined. This article demonstrates that a blood conservation protocol including low-dose anticoagulation and a restrictive transfusion strategy is

0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.09.031

ADULT CARDIAC

16.

patients having noncardiac thoracic operations. Ann Thorac Surg 2011;91:1674–80. Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin definition. JAMA 2012;307:2526–33. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130:461–70. Levey AS, Coresh J, Greene T, et al. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006;145:247–54. Fernandes CJ, Akamine N, De Marco FV, De Souza JA, Lagudis S, Knobel E. Red blood cell transfusion does not increase oxygen consumption in critically ill septic patients. Crit Care 2001;5:362–7. Walsh TS, McArdle F, McLellan SA, et al. Does the storage time of transfused red blood cells influence regional or global indexes of tissue oxygenation in anemic critically ill patients? Crit Care Med 2004;32:36–71. Napolitano LM, Corwin HL. Efficacy of blood transfusion in the critically ill: does age of blood make a difference? Crit Care Med 2004;32:594–5. Schmidt M, Tachon G, Devilliers C, et al. Blood oxygenation and decarboxylation determinants during venovenous ECMO for respiratory failure in adults. Intensive Care Med 2013;39:838–46. Shorr AF, Duh MS, Kelly KM, et al. Red blood cell transfusion and ventilator-associated pneumonia: a potential link? Crit Care Med 2004;32:666–74. Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med 2011;365:2453–62. Marik PE, Corwin HL. Acute lung injury following blood transfusion: expanding the definition. Crit Care Med 2008;36: 3080–4. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Intern Med 2012;157:49–58. Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med 2008;358:1229–39. Kor DJ, Kashyap R, Weiskopf RB, et al. Fresh red blood cell transfusion and short-term pulmonary, immunologic, and

AGERSTRAND ET AL BLOOD CONSERVATION IN ECMO FOR ARDS