Bein, Thomas*; Zimmermann, Markus*; Philipp, Alois†; Ramming, Michael*; Sinner, Barbara*; Schmid, Christof†; Müller, Thomas‡; Graf, Bernhard*; Schlitt, Hans Jürgen§; Weber-Carstens, Steffen¶
From the Departments of *Anesthesiology, †Cardiothoracic Surgery, ‡Internal Medicine, and §General Surgery, Regensburg University Hospital, Regensburg; and ¶Department of Anesthesiology and Operative Intensive Care, Universitätsmedizin Charitè, Berlin, Germany.
Submitted for consideration June 2010; accepted for publication in revised form February 2011.
Reprint Requests: Thomas Bein, PhD, Department of Anesthesiology, Regensburg University Hospital, Regensburg D-93042, Germany. Email: firstname.lastname@example.org.
The first two authors contributed equally to this work.
Pump-driven extracorporeal membrane oxygenation (ECMO) or pumpless arterio-venous interventional lung assist (iLA) is used in cases of life-threatening severe hypoxemic/hypercapnic respiratory failure.1–3 Advances in these techniques have reduced complication rates, and in a recent prospective randomized study, a reduction in mortality or severe disability was demonstrated in patients suffering from severe acute respiratory distress syndrome (ARDS) by the use of ECMO compared with patients managed by conventional mechanical ventilation.4 Although similar reports on improved survival with iLA are lacking, there are investigations on the beneficial effects of iLA on the extracorporeal removal of CO2 or the increase in arterial oxygenation.3,5 In contrast to ECMO, iLA is a single-use ultracompact extrapulmonary gas-exchange system perfused by the heart. A passive femo-femoral shunt flow generated by the arterial blood pressure through a lung assist device (Novalung GmbH, Talheim, Germany) produces a blood flow rate of approximately 1 L/min. Both ECMO and iLA are associated with possible complications, with frequencies ranging from 20% to 50%.3,6 The principal causes of complications are bleeding and thrombosis due to contact of the cellular components of the blood with the “non-biologic” surface of extracorporeal circuits, resulting in inflammatory and clotting responses.7 For the prevention of clotting and thrombosis, the surface of extracorporeal circuits and gas-exchange systems is coated with heparin, and the additional continuous infusion of heparin is advocated. Despite this anticoagulation management, the formation of clots resulting in increased resistance to blood flow is observed during iLA use. We reported that additional application of acetylsalicylic acid (ASA) in a patient treated with iLA for acute lung failure was helpful when a marked drop in blood flow rate in the extracorporeal system was observed despite adequate anticoagulation with heparin.8
In this prospective cohort study, we report our experiences with the additional use of low-dose ASA with heparin to prevent clotting activity. Our hypothesis was that this new anticoagulation management would help to prevent thrombosis without increasing bleeding complications.
Materials and Methods
Between December 2008 and January 2010, iLA was implemented in 15 patients suffering from severe respiratory failure (Table 1), and ASA was added to heparin for anticoagulation. The concept of evaluation, insertion, and clinical monitoring of iLA has been described in detail elsewhere.3 In brief, cannulae were inserted using Seldinger's technique in the femoral artery and in the contralateral femoral vein by experienced physicians. After connection of the gas-exchange membrane to the cannulae, the shunt flow was released and controlled continuously by transit time Doppler technology (Sono TT ultrasonic flowmeter, em-tec GmbH, Finning, Germany). After insertion of the iLA-system, a de-escalation of invasive ventilatory variables (i.e., tidal volume, inspiratory plateau pressure, frequency, and inspiratory fraction of oxygen) was performed to establish sufficient gas exchange by lung protective ventilation. Mean arterial blood pressure (MAP) was measured continuously during extracorporeal treatment, and volume and/or vasopressor agents were given to maintain MAP values ≥70 mm Hg.
The size of the arterial cannulation was individually selected to prevent ischemic complications of the lower limb: after ultrasonographic identification and assessment of the femoral artery and vein, an appropriately sized cannula that ensured sufficient peripheral blood flow (residual lumen ≥30%) was inserted. In most cases, the diameter of the arterial cannula was 15 French (Fr), whereas the venous cannula was selected to be 2 Fr larger, so that the flow resistance is not compromised. A continuous infusion of heparin ([Heparin-Natrium-25,000, Ratiopharm, Ulm, Germany] 200–600 IU/h) was given into the arterial cannula before the gas-exchange membrane to achieve a mild prolongation of the partial thromboplastin time (PTT ≈ 50 seconds). In addition, ASA (Aspisol, Bayer Vital, Germany) was administered intravenously after initiation of iLA with a dosage of 1.5 mg/kg actual body weight/d. We decided to apply ASA intravenously and not orally because critically ill patients often present with disturbed intestinal function and altered gastrointestinal motility and/or resorption capacity. The necessity for change of the artificial membrane was seen in situations, in which a marked reduction in iLA blood flow was observed (<0.9 L/min) despite hemodynamic stability (MAP ≥70 mm Hg) and without evidence for other reasons of reduced flow (kinking of cannula, etc.).
Two patients with acute bleeding problems (PTT >60 seconds and/or platelet count <60,000) and one patient with severe brain injury (Glasgow Coma Scale Score <9 and/or pathologic brain computed tomography scan) were excluded from ASA application. One patient with a suspected diagnosis of heparin-induced thrombocytopenia (HIT) was supplied with iLA but was excluded from the present analysis. Patients suffering from acute shock syndrome (norepinephrine infusion >0.4 μg/kg/min to maintain MAP ≥70 mm Hg) were excluded from iLA treatment and were, in cases of life-threatening hypoxemia/hypercapnia, connected to pump-driven ECMO. The transfusion of blood components followed our institution's policy aimed at ensuring a hemoglobin level of ≥9 g/dl, whereas fresh-frozen plasma was given in cases of active bleeding and/or a prothrombin time-international normalized ratio (PT-INR) >1.2. Platelets were transfused in situations of platelet blood levels below 60,000 counts/μl.
During iLA treatment, changes in coagulation parameters, free hemoglobin levels, the demand for transfusion of blood components, and hemodynamic variables were documented. Furthermore, we compared the function of the artificial membranes (oxygen transfer and capacity of CO2 removal) of the ASA group with that of a matched-pair control group (gender, diagnosis, age, and severity of disease) treated with heparin alone, which was drawn from our retrospective analysis in 90 patients with no ASA application.3 Our institutional review board denied the necessity for approval to publicize the data.
The statistical analysis was performed using SPSS software, version 18.0 (SPSS Inc., Chicago, IL). Most datasets are presented as means and standard deviations. Some datasets significantly varied from the pattern expected if they were drawn from a normal distribution, as revealed by the Kolmogorov-Smirnov method, and are presented as median and min-max values. Significance was analyzed by the Wilcoxon Signed Rank Test (intragroup analysis). Results were considered significant at p < 0.05.
All patients suffered from severe lung injury as assessed by Lung Injury Score and presented with general critical illness (Table 2). The mean duration of iLA treatment was 6.6 ± 3.7 days, and patients were mechanically ventilated for 14.3 ± 9.7 days. One patient died during hospital treatment due to multiple organ failure.
We found a significant increase in platelet count during iLA treatment, whereas other coagulation parameters remained stable. The hemoglobin level showed a moderate but statistically significant reduction on the 3rd day after initiation of iLA treatment (Table 3). The level of free hemoglobin was not reduced during iLA treatment and ASA application. The dosage of anticoagulation drugs required and the demand for blood transfusion are documented in Table 4. There was no increased bleeding activity or any demand for transfusion of red blood cell units, fresh-frozen plasma, or platelets resulting from ASA application in comparison with the matched-pair control group. Two examples of electron micrograph are presented in Figure 1, demonstrating iLA microfibers after the use of (Figure 1A) and without the addition of (Figure 1B) ASA. We saw no bleeding complications during or after the withdrawal of cannulae or in other organ systems. Hemodynamic variables are listed in Table 5. During iLA treatment and application of our combined anticoagulation management, the patients remained stable, and no increased demand for vasopressor infusion was noticed. iLA flow was significantly increased in the control group because of our former practice to use cannulae with a larger diameter (17 Fr) in comparison with our present practice (15 Fr). In addition, no transient ischemia was observed in the lower limb in which the femoral artery was cannulated.
Pulmonary gas exchange, and the amount of oxygen transfer and CO2 removal immediately after implementation of iLA and on days 3 and 5 are demonstrated in Table 6. Pulmonary gas exchange was significantly reduced in the control group in comparison with the ASA group at day 5 after implementation of iLA. The amount of O2 transfer tended to be increased during treatment in the ASA group but not in the control group. Relative changes of O2 transfer and CO2 removal expressed as a percentage on day 3 are shown in Figure 2. The relative changes of oxygen-transfer capacity were increased in the ASA group on day 3, but reduced in the control group, but for the changes in CO2 removal, we observed vice versa results. The use of membranes per patient (Table 7: membrane/patient ratio) tended to be decreased in patients treated with ASA (1.12 ± 0.34) in comparison with control (1.33 ± 0.62, p = 0.157). The incidence of iLA-associated complications was not different between groups. In the ASA group, one patient died due to multiple organ failure, whereas in the control group, five patients died, but initial SOFA and SAPS scores were higher in this group.
Recent advances in novel techniques of extracorporeal lung support (miniaturization of pump-driven ECMO and pumpless extracorporeal lung assist9) have stimulated the integration of these measures in a clinical regime for the management of severe acute respiratory failure10 aimed at providing adequate gas exchange and simultaneously allowing clinicians to prioritize lung-protective ventilation.
For both the ECMO and iLA systems, the extracorporeal gas-exchange membrane constitutes the core of the system, in which blood flows around bundles of gas-bearing hollow fibers with a large surface area. The gas exchange takes place at the boundary layer according to partial pressure gradients. With ECMO, the driving pressure is performed by a pump with a veno-venous cannulation, but the principle of iLA is the establishment of an arterio-venous pressure gradient in which the patient's heart acts as the pump. A common phenomenon of both techniques is the induction of an inflammatory and clotting response due to the contact of blood and its related components with the nonbiologic surface of the extracorporeal circuit. This response leads to the consumption and activation of procoagulant and anticoagulant components of cellular and noncellular cascades. In a recent autopsy series of ECMO patients, unexpectedly high rates of systemic thromboembolic events with an almost linear increase with duration of extracorporeal support (80% incidence after 10 days ECMO use) were found.11 Therefore, adequate anticoagulation is crucial for the effective use of extracorporeal systems while, on the other hand, possibly increasing the risk of bleeding. Consequently, the goal of anticoagulation for extracorporeal lung support is to prevent thrombosis in the patient/technical device and prevent excessive bleeding. Current recommendations are based on the continuous infusion of heparin monitored by various coagulation parameters (i.e., activated clotting time, PTT, and PT-INR.7).
Blood flow by a pumpless iLA device was identified as the main determinant of efficacy of oxygenation improvement and CO2 elimination.12 Conversely, it has been shown by electron and fluorescence microscopy that cellular deposits are frequently observed on the surface (polymethylpentene) of gas-exchange membranes, increasing the resistance to blood flow and affecting the gas-exchange capacity.13
We, therefore, decided to add an inhibitor of platelet aggregation to heparin, and in the present investigation, we present a feasibility study. The main results of our prospective cohort study are the following: 1) addition of ASA (1.5 mg/kg/d) does not increase the risk of bleeding or the amount of transfused blood components compared with “conventional” use of heparin alone; 2) laboratory values of coagulation parameters were unaffected by the addition of ASA to heparin in our patients; and 3) we found evidence that additional ASA preserved the amount of oxygen transfer through the membrane in comparison with a matched-pair group treated with heparin alone. Furthermore, we observed a trend for a lower membrane/patient ratio, which allows for a less necessity of changing the membrane.
Most of our patients were trauma patients suffering from severe posttraumatic ARDS due to direct or indirect involvement of the lung. Although it has been shown in a series of 28 patients that extracorporeal life support has been safely used after trauma,14 the indication for ECMO or iLA is critical in these patients because the risk for bleeding complications is high, and conversely, trauma is known to promptly induce coagulation cascades. The analysis of seven trauma patients in this study indicates that the use of low-dose ASA is safe and effective, but we do not recommend the use of ASA in patients with moderate-to-severe traumatic brain injury or severe preexisting coagulation disorder. The “ideal” anticoagulation management of patients on extracorporeal lung support should be studied in future experimental and clinical studies. We believe that the results of this study may impact additional investigations on the use of low-dose thrombocyte aggregation inhibitors to optimize anticoagulation without harming the patient.
1. Schuerer DJ, Kolovos NS, Boyd KV, Coopersmith CM: Extracorporeal membrane oxygenation: Current clinical practice, coding, and reimbursement. Chest 134: 179–184, 2008.
2. Brogan TV, Thiagarajan RR, Rycus PT, et al: Extracorporeal membrane oxygenation in adults with severe respiratory failure: A multi-center database. Intensive Care Med 35: 2105–2114, 2009.
3. Bein T, Weber F, Philipp A, et al: A new pumpless extracorporeal interventional lung assist in critical hypoxemia/hypercapnia. Crit Care Med 34: 1372–1377, 2006.
4. Peek GJ, Mugford M, Tiruvoipati R, et al; for CESAR trial collaboration: Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): A multicentre randomised controlled trial. Lancet 374: 1351–1363, 2009.
5. Zimmermann M, Bein T, Arlt M, et al: Pumpless extracorporeal interventional lung assist in patients with acute respiratory distress syndrome: A prospective pilot study. Crit Care 13: R10, 2009.
6. Hemmila M, Rowe S, Boules TN, et al: Extracorporeal life support for severe acute respiratory distress syndrome in adults. Ann Surg 240: 595–605; discussion 605–607, 2004.
7. Oliver WC: Anticoagulation and coagulation management for ECMO. Semin Cardiothorac Vasc Anesth 13: 154–175, 2009.
8. Philipp A, Müller T, Bein T, et al: Inhibition of thrombocyte aggregation during extracorporeal lung assist: A case report. Perfusion 22: 293–297, 2007.
9. Kopp R, Bensberg R, Henzler D, et al: Hemocompability of a miniaturized extracorporeal oxygenation and a pumpless interventional lung assist in experimental lung injury. Artif Organs 34: 13–21, 2009.
10. Deja M, Hommel M, Weber-Carstens S, et al: Evidence-based therapy of severe acute respiratory distress syndrome: An algorithm-guided approach. J Int Med Res 36: 211–221, 2008.
11. Rastan AJ, Lachmann N, Walther T, et al: Autopsy findings in pastients on postcardiotomy extracorporeal membrane oxygenation (ECMO). In J Artif Organs 29: 1121–1131, 2006.
12. Müller T, Lubnow M, Philipp A, et al: Extracorporeal pumpless interventional lung assist in clinical practice: Determinants of efficacy. Eur Respir J 33: 551–558, 2009.
13. Lehle K, Philipp A, Gleich O, et al: Efficiency in extracorporeal membrane oxygenation—Cellular deposits on polymethylpentene membranes increase resistance to blood flow and reduce gas exchange capacity. ASAIO J 54: 612–617, 2008.
14. Cordell-Smith JA, Roberts N, Peek GJ, Firmin RK: Traumatic lung injury treated by extracorporeal membrane oxygenation (ECMO). Injury 37: 29–32, 2006.