| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiothoracic Surgery, University Hospital, Regensburg, Germany
b Department of Internal Medicine II, University Hospital, Regensburg, Germany
c Department of Internal Medicine I, University Hospital, Regensburg, Germany
Accepted for publication April 7, 2008.
* Address correspondence to Dr Flörchinger, Department of Cardiothoracic Surgery, University Hospital, Franz-Josef-Strauss-Allee 11, Regensburg, 93053, Germany (Email: christof.schmid{at}klinik.uni-regensburg.de).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
| This article has been selected for the open discussion forum on the CTSNet Web Site: http://www.ctsnet.org/sections/newsandviews/discussions/index.html
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Since 1996, 159 patients with an age ranging from 7 to 78 years were provided with a PECLA system. Fifteen patients were referred to us by air or ground transport after insertion of the system in a peripheral hospital.
Results: Main underlying lung diseases were acute respiratory distress syndrome (70.4%) and pneumonia (28.3%). Pumpless extracorporeal lung assist lasted for 0.1 to 33 days, mean 7.0 ± 6.2 days; cumulative experience was greater than 1,300 days. Successful weaning and survival to hospital discharge was achieved in 33.1% of patients after a mean PECLA support of 8.5 ± 6.3 days. During PECLA therapy, 48.7% of patients died, mainly as a result of multiorgan failure after a mean interval of 4.8 ± 5.1 days. Inability to stabilize pulmonary function was noted in 3% of patients only. After PECLA, 30-day mortality was 13.6%. In a subgroup analysis, best outcome was obtained in patients after trauma.
Conclusions: Pumpless extracorporeal lung assist is a simple and efficient method to support patients with deteriorating gas exchange for prolonged periods to allow the lung protective ventilation and transportation. Best indication for use of PECLA is severe hypercapnia and moderate hypoxia.
Acute lung failure is associated with high mortality rates ranging between 35% and 50% [1]. Most common causes of lung failure are pneumonia, septicemia, and polytrauma, ultimately resulting in acute respiratory distress syndrome (ARDS). Clinically, acute lung failure is characterized by respiratory insufficiency with severe hypoxemia or hypercapnia. The treatment strategy depends on the knowledge of the underlying disease. If the latter is known, as for example in pneumonia or sepsis, the focus of infection is eradicated. Lung-protective ventilation with adjusted positive end-expiratory pressure remains the most effective therapeutic tool [1–3]. Kinetic therapy, inhalation of nitric oxide, and use of steroids are further therapeutic options, but not evidence based.
After the first clinical use of extracorporeal membrane oxygenation (ECMO) in 1972 [4], there has been a continuous improvement of the technique of artificial lung support. Currently, ECMO is a routine procedure with a 35-year experience, usually carried out in patients with severe ARDS as a rescue therapy. Unfortunately, ECMO therapy often has to be restricted with regard to support duration for its negative side effects, such as hemolysis, coagulation disorders, and technical complications.
At the University Hospital of Regensburg, Germany, a pumpless extracorporeal lung assist device (PECLA) was developed by an interdisciplinary group of perfusionists, cardiac surgeons, internists, and anesthesiologists in cooperation with the industry (Novalung GmbH, Hechingen, Germany). An oxygenator with a low flow resistance was connected to cannulas in arterial and venous vessels as well as to an external oxygen supply. The first clinical use took place in 1996 [5]. With the introduction of this minimized pumpless lung assist system using the blood pressure gradient between arterial and venous circulation as the driving force, a reduction of the negative side effects, eg, blood trauma and coagulation disorders and the high costs of pump-driven membrane oxygenation, could be achieved with a comparable mortality to ECMO therapy in selected patient groups [6]. We report on our 10-year experience in PECLA in patients with acute lung failure.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
Before insertion of a PECLA system, a clinical assessment of the hemodynamic status of the patient was performed, including placement of a Swan-Ganz catheter or echocardiography. As the PECLA is a pumpless device, a cardiac index greater than 3 L · min–1 · m–2 and a mean arterial blood pressure of greater than 70 mm Hg was considered necessary because an arteriovenous shunt volume of 1.0 to 2.5 L/min has to be tolerated by the patient's circulation. If these requirements could not be met with or without catecholamine support (epinephrine, norepinephrine) a pump-driven device was used instead.
Pumpless extracorporeal lung assist therapy was terminated with a PaO 2 greater than 80 mm Hg and FiO 2 less than 0.45 for more than 24 hours (weaning criteria).
Pumpless Extracorporeal Lung Assist System and Its Placement
The main component of the PECLA is a membrane oxygenator (iLA Membrane Ventilator, Novalung GmbH, Hechingen, Germany) with a low resistance (of approximately 15 mm Hg at 2.5 L/min blood flow) between inflow and outflow. The diffusion membrane of the oxygenator is manufactured of poly-4-methyl-1-penten and has a surface of 1.3 m2 to which an oxygen flow of 1 to 12 L/min can be administered. The system is primed with 250 to 300 mL of Ringer's lactate solution. The cannulas for arterial insertion were sized from 15F to 19F, those for venous insertion from 17F to 19F. The PECLA system as well as the tubing was coated with high-molecular-weight heparin, which allows minimizing anticoagulation and surveillance of it with the activated clotting time (target range, 130 to 150 seconds). Further components include an ultrasound sensor (Transonic Inc, Ithaca, NY) for continuous flow measurement as well as the arterial and venous lines connected to the respective insertion cannulas (Fig 1).
|
|
Measured Variables
Inspiratory oxygen fraction, PaO
2, and PaCO
2, as well as arterial oxygen saturation, serum lactate, and blood pH levels, were recorded immediately before PECLA implantation, 2 and 24 hours after implantation, and before and 24 hours after terminating PECLA therapy. Likewise, respiratory variables including respiratory rate, tidal volume, and mean airway pressure were noted. Coagulation variables, serum creatinine, liver enzymes, and hematologic variables were analyzed daily.
Statistical Analysis
Statistical analyses were performed with Sigma-Stat 3.1 (Scientific Solutions Inc, Lausanne, Switzerland) and SPSS (SPSS Inc, Chicago, IL). Differences between survivors and nonsurvivors were calculated by Mann-Whitney-Wilcoxon test.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Seventeen patients had been primarily provided with a pump-driven venoarterial ECMO system owing to concomitant low cardiac output. With recovery of sufficient myocardial pump function, the ECMO system was switched to a PECLA system by removing the pump from the circuit.
In 15 cases, PECLA therapy was established in a peripheral hospital by an anesthesiologist or cardiac surgeon and a perfusionist experienced with PECLA technique before transfer to our institution. Transport of these patients was accomplished by helicopter (n = 12) or ambulance car (n = 3).
Pumpless Extracorporeal Lung Assist Data
Mean PECLA support interval was 7.0 ± 6.2 days (range, 0 to 33 days) with a cumulative experience of 1,308 days. A total of 214 membrane oxygenators was used, which corresponds to 1.3 per patient. The predominant size of the arterial cannula was 17F (86 patients, 51.2%), whereas for venous access 19F (67 patients, 40.3%) and 21F (60 patients, 38.4%) cannulas were preferred. Percutaneous femoral cannula placement was successful in all but 6 patients who required an open surgical insertion (femoral artery, 5 patients; femoral vein, 1 patient).
The arteriovenous shunting blood flow over the PECLA varied from 1.9 to 2.1 L/min, with a delivered oxygen flow of 6 to 13 L/min.
Main reason for oxygenator exchange was thrombus formation despite an assumed sufficient anticoagulation (27 patients), and retention of air bubbles (22 patients). Plasma leakage was rare and occurred in 7 patients only. In these patients, a membrane oxygenator with a membrane of microporous polypropylene was used. Thrombosis of the entire system with consecutive no-flow developed in 8 patients. Four of them were inadequately anticoagulated, whereas 2 patients each suffered from heparin induced thrombocytopenia type II and device failure, respectively.
Gas Exchange and Hemodynamics
A significant improvement of oxygenation was notable during PECLA support in all but 3 patients (98%). Before PECLA mean FiO
2 was 0.96 ± 0.09, which steadily improved to an FiO
2 of 0.48 ± 0.13 (p = 0.001) before terminating PECLA and resuming conventional mechanical ventilator therapy. Likewise, PaO
2 continuously increased under PECLA therapy from 66 ± 24 mm Hg before initiation to 79 ± 19 mm Hg after 24 hours, and 91 ± 17 mm Hg before removal of the cannula (p = 0.001). After PECLA explantation, oxygenation remained stable (FiO
2 = 0.5 ± 0.13, not significant). A similar significant improvement was noted with regard to PaCO
2, which decreased from 67 ± 24 mm Hg before PECLA therapy to 35 ± 7 mm Hg after only 24 hours of PECLA therapy (p = 0.001). The normalization of carbon dioxide levels remained stable until PECLA therapy was stopped (39 ± 17 mm Hg; Fig 3;
Table 3). After improvement of respiratory function and termination of PECLA, patients underwent further standard ventilator therapy for 16.6 ± 15.7 days.
|
|
|
|
|
Ischemia of the lower limb was noted in 13 patients. In these cases the arterial cannula was either exchanged with a smaller one or moved to the contralateral femoral vessel. A compartment syndrome with a necessary fasciotomy occurred in 4 patients. One patient with this complication finally underwent lower leg amputation.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The aim of PECLA insertion is to allow lung-protective ventilation and to improve gas exchange. Thus, native lung function is supported, and the diseased lung may recover better as artificial ventilation can be downgraded. Accordingly, additional iatrogenic lung injury such as barotrauma and volutrauma caused by mechanical ventilation with high tidal volumes and high peak inspiratory pressures can be reduced [11, 12]. Our results demonstrate that moderate transfer of oxygen and efficient elimination of carbon dioxide is well achieved in most cases. The latter is also supported by the course of the serum lactate levels during PECLA, which dropped to low levels in the long term.
Criteria for PECLA placement were in accordance with the fast entry criteria of the Extracorporeal Life Support Organization [13]. In either situation all conservative means to improve gas exchange should be tried at first, but in case of persistent moderate hypoxia and severe hypercapnia despite aggressive mechanical ventilation therapy, indication for PECLA placement may be established.
Apart from PaO 2 and PaCO 2, the left ventricular function has to be considered also when establishing an indication for PECLA. In the early days of PECLA, the system was used as a rescue procedure, also in patients with considerably impaired myocardial pump function (instead of using a venovenous ECMO). During the years, increasing experience has shown that the artificial arteriovenous shunt may not be hemodynamically tolerated. Accordingly, hemodynamic judgment before PECLA placement is necessary—for that purpose, a Swan-Ganz catheterization is beneficial and was performed in most cases (96 patients). Alternatively, echocardiography can be performed to assess myocardial contractility and to estimate cardiac index.
If cardiac index is greater than 3 L · min–1 · m–2, PECLA may be placed. If inadequate cardiac pump function is noted, PECLA should not be used as ECMO is a far better alternative.
If pump function worsens after PECLA implantation, an additional pump can be included into the circuit and venoarterial ECMO instituted (with flow reversal through the cannulas). In those cases placement of the arterial cannula in the right subclavian artery seems to be good alternative to assure perfusion of supraaortic vessels with oxygenated blood.
Vascular access can be most easily gained at the femoral site. The femoral artery is usually easily palpated, whereas the femoral vein can be punctured. As the patients are in a critical condition, femoral cannulation hindering mobilization is not an issue. In patients with severe calcifications owing to peripheral arterial occlusive disease in whom it is impossible to place a femoral artery cannula, we switch to the right subclavian artery and use a small tubular Dacron prosthesis and open surgical exposure. Access to the right subclavian vein is possible, too. Axillary vessels as a safe cannulation site for either direct cannulation or end-to-side graft interposition have also been used in an experimental setting in animals by Iglesias and colleagues [14], but this approach has not been performed in humans so far. Because of cannula placement in an arterial vessel, PECLA systems are more susceptible to vascular and bleeding complications in contrast to a venovenous ECMO. In our experience, however, most complications can be avoided by careful assessment of the insertion site of the vessels by ultrasound and by choosing an appropriate cannula.
Anticoagulation during PECLA is still a matter of discussion. Because the whole system including tubing is heparin-coated, full anticoagulation is unnecessary. Moreover, as the patients are usually in a detrimental condition with multiorgan dysfunction or failure, the coagulation system is frequently impaired. As a consequence, we perform anticoagulation assessed by activated partial thromboplastin time (target range, 50 to 60 seconds) as well as a mild antiaggregation with aspirin to prevent oxygenator thrombosis [15]. If the clinical situation demands heparin withdrawal, we liberally stop it for 2 to 3 days, as the oxygenator can easily be replaced in case of thrombosis.
The PECLA offers feasibility for further applications. Fischer and colleagues [16] have reported on PECLA therapy in 12 patients with severe lung failure awaiting lung transplantation. We used this in 2 cases as well. Another patient (not included in this study) with pulmonary hypertension and acute lung failure was bridged to lung transplantation with a PECLA device implanted as a pulmonary artery to left atrium shunt [17]. Brederlau and associates [18] studied in an animal model the combination of PECLA with high-frequency oscillatory ventilation in a porcine model of lavage-induced acute lung injury. However, PECLA is not only a valuable tool for intensive care units, but also for transportation of critical patients. As previously reported, patients with severe ARDS after trauma have been transferred to a primary care hospital with the aid of a PECLA system [19]. Owing to its simple design, PECLA is more suitable for interhospital transfer as compared with ECMO as an accompanying pump technician is not required, and only minor training for ambulance personnel is necessary.
Pumpless extracorporeal lung assist is not the only means for paracorporeal lung support. Wang and coworkers [20] are focusing their interest on the development of a so-called OxyRVAD, a combination of oxygenator and right ventricular assist device, that allows ambulation of the patients. In animal experiments, their device demonstrated a blood flow through the system of about 3 L, a carbon dioxide removal of 200 ± 19 mL/min, and an oxygen transfer of 144 ± 44 mL/min. Eash and colleagues [21] are developing an intravenous respiratory assist catheter, which uses microporous hollow fiber membranes wrapped into a bundle around a pulsating balloon. They achieved maximum pulsation carbon dioxide gas exchange rates of 174.4 ± 1.8 mL · min–1 · m–2.
Compared with recent results in ARDS patients treated with pump-driven ECMO or conventional respirator therapy, mortality rate in this report appears to be high [22, 23]. However, this study reflects a 10-year single center experience of 159 consecutive patients who were provided with PECLA support. Unlike other previous studies, all patients supported by PECLA were included, ie, we also included patients who died within 24 hours unrelated to respiratory failure. Moreover, an institutional learning curve with regard to efficacy of PECLA in terms of oxygenation, and carbon dioxide removal as well as handling of this device has to be considered. Indication for PECLA use has changed fundamentally from the first attempts of ARDS treatment to its routine application now. In the beginning, PECLA was considered a rescue procedure regardless of circulatory depression. As a consequence, we lost 13 patients owing to low cardiac output while on PECLA support. Meanwhile, we established an algorithm for extracorporeal respiratory support. In case of severe hypoxia or circulatory depression a pump-driven ECMO is the method of choice for its better outcome. The indication for PECLA is limited to patients with severe hypercapnia, respiratory acidosis, and accompanying moderate hypoxia with unimpaired left ventricular function. This patient cohort is mainly represented by trauma patients and promises low mortality and appropriate outcome with PECLA therapy. Further prospective and randomized studies are necessary to evaluate the beneficial effects of PECLA support in defined patient subpopulations.
Pumpless extracorporeal lung assist is a simple and efficient method to support patients with deteriorating gas exchange for prolonged periods. Elimination of carbon dioxide is more effective than oxygenation. Thus, indications for PECLA are severe hypercapnia and respiratory acidosis with moderate hypoxia in patients with normal cardiac output. Pumpless extracorporeal lung assist can be used in intensive care units, but it also alleviates transportation in patients with severely compromised pulmonary function.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Starting in the '90s through the present, our laboratory explored the concept of CO2 normalization with the development of percutaneous arteriovenous CO2 removal. Our studies continue to show that arteriovenous CO2 removal ameliorates the pathophysiology of acute respiratory distress syndrome (ARDS).
The group from Regensburg was among the first to treat patients with percutaneous arteriovenous CO2 removal, which they term PECLA. As pioneers, they experienced a steep learning curve regarding vascular access, anticoagulation dosage, patient selection, and critical care management. The data presented today is seductive. This single-institution experience clearly shows feasibility and implies proof of concept. Feasibility has been established by the evolution of the technology and techniques described. Proof of concept is supported by the apparent impact of the technique upon the disease and outcomes. Likewise, the results continue to parallel our large animal studies.
The weaknesses of this 10-year single-institution study are highlighted by the evolution of the technology and an evolving understanding and management of ARDS. Over the time span of this study, techniques of vascular access, circuit components, and critical care management have improved. Likewise, midway through the experience, low tidal volume ventilation management techniques became standard.
Finally, patient selection once again appears to be a major determinant of outcomes. Defining the onset of primary and secondary ARDS continues to elude investigators. My question for the investigators is, given your 10-year experience, what patient population would you select for a prospective randomized multicenter outcome study that is clearly called for by this experience? What training or experience would you require to qualify a center for participation in such a trial?
I congratulate the investigators for such a pioneering effort.
DR FLÖRCHINGER: Thank you for your comments. I think the best population for a prospective study is patients after multiple trauma. They are the best suitable patients because they have a normal cardiac output, and these are the patients with the best outcome. A special training for participating centers would be minimal.
DR ZWISCHENBERGER: You realize what he is saying; take a trauma patient that appears to be dying of ARDS, then counterintuitive to the fact that the patient is hypoxic and dying of respiratory failure, you insert percutaneous cannulas, anticoagulate the patient, and target CO2 normalization. This approach requires a thorough understanding of the relationship between CO2 normalization and gas exchange. Arteriovenous CO2 removal has clearly been shown to improve the ventilation-perfusion (V/Q) matching and improve the pathophysiology of ARDS in large animals. The most significant finding in this whole study is the fact that CO2 normalization has such a large impact on the outcomes of ARDS in select patients as well.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Puehler, A. Philipp, and C. Schmid Paracorporeal artificial lung circuit as a possibility for bridge to lung transplantation. Ann. Thorac. Surg., July 1, 2009; 88(1): 352 - 352. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
