Acute respiratory distress syndrome (ARDS) is a pathology that is commonly encountered in intensive care, and the resulting mortality rate remains elevated despite recent progress in mechanical ventilation and life support practices.1–4 In ARDS patients with no associated medical problems, the appearance of organ failure followed by multiorgan system failure is a sign of a fatal condition.5 In the long term, ARDS survivors have moderate alterations in respiratory function but primarily have extra-pulmonary manifestations (e.g., neuromuscular and cognitive manifestations).6 These extra-pulmonary elements suggest a possible iatrogenic cause (e.g., paralysis, immobility) or other problems that are directly linked to acute hypoxia.
Since the 1970s, the use of extracorporeal membrane oxygenation (ECMO) has been considered as an alternative therapy for ARDS.7 ECMO allows the lungs to rest while limiting ventilation in terms of volume and respiratory rate and ensuring gas exchange. This homeostasis can potentially prevent ischemia-induced organ failure. In 2009, the first multicenter study on the modern management of ARDS, with or without ECMO,8 was published. This study showed a significant survival benefit in the ECMO group. That same year, the use of ECMO for ARDS was expanded to address the respiratory difficulties caused by the avian flu (H1N1) epidemic.9
The venovenous ECMO circuit consists of a venous inflow route, a pump that allows blood to circulate through the circuit, an oxygenation membrane connected to a gas inlet, and an “arterial” outflow that reintroduces oxygenated blood to the right atrium. There is generally double access via a jugular catheter for outflow and a femoral catheter for inflow to the pump. Other types of access are possible in difficult cases, including triple access, which allows increased blood flow to the pump.9 The main problems in ECMO are related to the multiple access points, which increase local complications (e.g., skin incisions) and limit patient mobility. In addition, there is a possibility of closed-loop blood circulation in the ECMO circuit (recirculation of oxygenated outflow that reenters the inflow cannula). This recirculation reduces the efficiency of ECMO while decreasing the real volume that exits the ECMO circuit and flows to the right heart, therefore entering the systemic circulation. A bicaval double-lumen cannula (Avalon©, Avalon Laboratory, Los Angeles, CA) allows a single route of insertion and ejection through the same port. After insertion through a jugular route, the cannula descends into the inferior vena cava. Aspiration occurs through the distal end in the inferior vena cava and through openings higher up the cannula, in the superior vena cava. Ejection occurs between the aspiration openings, which sit directly opposite the right atrium; the blood flows toward the tricuspid valve. The advantages are a single venipuncture and the possibility of less restricted mobility.10 A combined system of admission/ejection could further reduce the risk of recirculation.11 The first use of this cannula was reported in 2010 in 11 patients with respiratory failure from diverse causes.12 Apart from the cannulation, the ECMO circuit was the same for all of the patients. Cannulas between 27 and 31-Fr. were used for a median of 78 hours with rates of 3.8 and 4.3 L/minute, respectively. The cannulas were inserted under transthoracic echocardiography (TTE) monitoring. No deaths were attributable to the use of cannulas. During use, 3 cannula displacements occurred (1 in the right atrium, causing significant recirculation, and 2 in the hepatic vein) that required simple remobilization, along with 1 cannula thrombosis after anticoagulation failure. After an 18-month period of commercial availability, we surveyed French practitioners about their experiences with this cannula.
Materials and Methods
From March 2011 to April 2011, we contacted all of the French centers that had used a dual-lumen cannula in an adult at least once. These centers were identified from the data of the company in charge of the distribution of the cannula in France. The reporting time began with the start of French commercialization of the cannula (September 2009) and ended with the data collection period. The retrospective data described patient demographics and cannula use. In all centers, there was one questionnaire for each use of a cannula. Apart from the cannula, the material that was used for the ECMO circuit and the management of the material were not specified. The questionnaire consisted of 3 sections. Details regarding the insertion, cannula size, operator, whether a perfusionist was present, and the time of insertion and monitoring were reported. The data on performance included the duration of use, maximum flow, and extubation or mobilization (e.g., chair, lounge chair, prone). Tolerance was evaluated by the presence of hemolysis (quantitatively estimated by the patient’s primary physician, principally based on the blood transfusion requirement or decrease in hemoglobin concentration with no other explanation, such as external hemorrhage), local or systemic complications, incidence of thrombosis, the need to remobilize, and the need to adjust the cannula for multiple access point capability. In addition, we investigated the imaging methods used for cannula monitoring and the different events that were detected during the course of that imaging. Finally, at the time of removal, we requested the identity of the operator, the method of hemostasis, and the state of the cannula. This study was approved by the ethics committee of our institution.
The quantitative data are expressed as mean ± standard deviation. Analyses were performed using Student’s t-test or analysis of variance (ANOVA). The qualitative data are expressed as percentages, and they were compared using a χ2 test. Statistical significance was set at p < 0.05.
By the end of the study period, 52 instances of dual-lumen cannula use were reported from 11 university hospitals and 7 general hospitals. Only 1 center refused to collect data. The mean cannula use was 3 cannulas per center; at the extremes, 7 centers used 1 cannula each, and 1 center used 9 cannulas. The patient demographic data are summarized in Table 1. The mean patient age was 45.3 ± 17.8 years. Global mortality was 46% (n = 24), 23% (n = 12) of whom died during supportive care with the cannula (3 cerebrovascular accidents, 1 air embolus, and 8 multiorgan system failures). The primary indication for the use of supportive care was ARDS (77%). The other patients included 7 hypercapnic failures in patients with cystic fibrosis before lung transplantation, 2 pulmonary edemas after thoracic surgery, 1 pulmonary embolism, 1 major bronchopulmonary fistula, and 1 asphyxia after tracheal stenosis. Only the patient with pulmonary embolism, 1 patient with pulmonary edema, and the patient with bronchopulmonary fistula did not survive.
The cannula was placed by a surgeon in 27 cases (52%), by an intensivist in 12 cases (23%), and by a multidisciplinary team in 13 cases (25%). All physicians inserted the cannulas percutaneously. The times required for cannula placement were 26 ± 13, 25 ± 16, and 31 ± 13 minutes, respectively, for each type of operator (p = 0.4). Cannula placement was performed under transesophageal echocardiography (TEE) guidance in 67% of cases (n = 35), TTE guidance in 25% of cases (n = 13), fluoroscopy in 4% of cases (n = 2), and with no image guidance in 4% of cases (n = 2). A perfusionist assisted with the cannula placement in 34 cases (65%) (Table 2).
All of the assistance devices were operated in conjunction with anticoagulation therapy with unfractionated heparin. No cases of cannula thrombosis were reported. The mean duration of supportive care with the dual-lumen cannula was 8 ± 7 days (maximum 32 days, median 7 days [5;7]). There was no difference in the length of use as a function of the cannula size (p = 0.33). The maximum observed flow was 2,175 ± 556 ml/minute for 20 Fr. (n = 4), 3,207 ± 653 ml/minute for 23 Fr. (n = 14), 3,693 ± 729 ml/minute for 27 Fr. (n = 25), and 5,490 ± 984 ml/minute for 31 Fr. (n = 9) (Figure 1). These rates were equivalent to those reported by the manufacturer for 20- and 27-Fr. cannulas and were higher than those predicted for 23- and 31-Fr. cannulas (in ascending order, the expected rates were 2,300, 2,750, 4,000, and 4,500 ml/minute). Daily monitoring of the cannula position was multimodal, involving chest x-rays (90%, n = 47), TEE (46%, n = 24), and TTE (42%, n = 22).
None of the cannulas were found to be abnormal upon removal. There was moderate hemolysis in 11 patients (21%). At the insertion site, there was 1 instance of inflammation, and bleeding was observed in 9 patients (light bleeding in 7, moderate bleeding in 1 (23 Fr.), significant bleeding in 1 (31-Fr.)). In 3 patients, the cannula was directly responsible for the complications. Among these patients, 1 developed a tricuspid valve lesion, and another had a tear in the right atrium that was complicated by tamponade. These 2 patients survived after surgical interventions. The third patient died after the cannula migrated into the right ventricle. For this patient, it is difficult to directly associate the problem with the cannula (the patient had unstable severe multiorgan failure and was worsening rapidly before catheterization). The other complications that were observed after placement, excluding those in patients who died during supportive care, are as follows: 1 superior vena cava thrombus, 1 jugular vein thrombus, 2 instances of acute renal failure, 2 incidences of septic shock, 1 case of hepatic cytolysis, and 1 case of multiorgan failure. In 9 patients, it was necessary to reposition the cannulas 11 times during supportive care. In 8 cases, vascular access was modified. In 2 cases, cardiac failure precipitated conversion to venoarterial access. In all other cases, double access was achieved to increase the flow in patients with persistent hypoxemia. For 5 of these patients, a femoral cannula was added to the dual-lumen cannula. Despite the fact that physicians did not collect data on the flow after these interventions, they considered the improvements sufficient to restore the clinical situations. In 3 other cases, right ventricular migration was the cause of cannula repositioning.
Five patients were extubated during supportive care. One patient was moved to a prone position, 3 sat on lounge chairs, and 5 sat in upright chairs.
In the patients who were removed from supportive care (n = 39), the cannulas were removed by a surgeon (24 patients), an intensivist (13 patients), or a collaborating surgeon and intensivist (2 patients). The intensivists, regardless of whether they worked with surgeons, achieved hemostasis by compression and eventually placed cutaneous sutures. For the patients whose cannulas were withdrawn by surgeons, 8 received sutures in the jugular vein, and 16 others received compression with cutaneous sutures. No complications were observed after cannula removal.
After introduction to the commercial market, more than 50 Avalon© dual-lumen cannulas have been used in France. Here, we report 52 experiences with this cannula. The primary indication for the use of cannula was ARDS with failure of mechanical ventilation, and the goal was to permit either oxygenation or the elimination of carbon dioxide. In general, most of the contacted teams seemed to be satisfied with this catheter. During its use, the observed flow rates were either equivalent or superior to the manufacturer’s predictions. The limitation of flow by the catheter configuration did not decrease its efficiency. In fact, the association of 2 lumens in a single cannula poses the problem of blood flow resistance. According to Poiseuille’s law, a reduction in the internal caliber induces a 4-fold increase in the resistance. Therefore, limiting the flow rate was the main concern. In the end, the flow rates were sufficient to achieve respiratory support in severely hypoxemic patients when compared with the flow rates obtained with multiple access methods.13 However, for patients in whom the flow was insufficient, the possibility of adding a supplementary cannula allowed the physician to return to a double-access situation (rather than triple access if there were already 2 ports in use). Given the low incidence of complications related to the size of the cannula, a large-caliber (27 or 31-Fr.) cannula should be the preferred size for respiratory support in hypoxemic adults to avoid the use of additional cannulas. This cannula size allows flow rates exceeding 4 L/minute, even reaching 5 L/minute for 31-Fr. cannulas.
During cannula placement and after controlling its position, there was no evidence of bleeding, vascular injury, or early false pathway. However, we cannot exclude that some of the complications that were experienced by our patients resulted from transient mispositioning of the cannula. Both the intensivists and surgeons were experienced in the percutaneous placement of central venous lines. The observed time difference for cannula placement between intensivists and surgeons is difficult to interpret. It is uncertain whether the procedure times were recorded in identical manners. On the other hand, assistance from a perfusionist did not decrease the placement time or the occurrence of complications. Similarly, after the cannula was withdrawn, regardless of the method of hemostasis (simple compression or jugular sutures) and the type of physician chosen to withdraw the cannula (intensivist or surgeon), there were no differences in the patients’ clinical courses.
A majority of the teams used TEE to control the cannula position during its insertion. This technique allows the physician to control the orientation of the ejection blood flow and the position of the ejection outlet in the vena cava. Most of the other teams used TTE. Although the latter technique cannot visualize the ejection flow, it still allows precise localization of the position of the distal end of the cannula in the inferior vena cava. In this situation, the absence of flux control does not seem to influence the flow or the efficiency of support. The situation is the same for cannulas that were positioned under fluoroscopy or without image guidance. Conversely, secondary displacements were observed, particularly in the right ventricle. This problem has been described previously.12 The failure of the cannula to descend into the inferior vena cava might be responsible for these observed displacements. In this study, patient height could have had an influence because, apart from 20 Fr., all of the cannulas are 29 cm in length. To avoid this influence, TTE could be the most accurate means of control because this technique allows the best visualization of the inferior vena cava and its contents, as described previously.14 In that regard, most teams used chest x-rays for daily position monitoring, which provided verification that a secondary displacement of the probe had not occurred. Ultrasound, transthoracic or transesophageal echocardiography, or a combination of techniques were used systematically.
The main complication reported was migration of the distal end of the cannula into the right ventricle. These episodes were detected either by the imaging methods previously described or by the indication of recirculation (sharp drop in the efficiency of supportive care). These migrations had direct and severe consequences in 2 patients who required surgical repair. These 2 patients were of average height (175 and 180 cm). The cannula placement was performed under TEE control only. These complications could suggest the importance of inserting the distal end deeper into the inferior vena cava. We report no migration into hepatic vessels, which has been described previously.12 Finally, the structure did not deform or cause internal thrombus, regardless of the duration of use (including in patients who died during supportive care). These results are in accordance with a recent single-center study that reported 26 experiences.15
Very few teams took advantage of the ability to mobilize patients during the course of their supportive care. Moving 3 patients to lounge chairs and 5 others to upright chairs was likely to be advantageous. A strict supine position is required for a femoral cannula. A prone position was achieved in 1 patient without encountering complications, proven this position is possible with double access.16
Our study has several limitations. First, the data collection was performed retrospectively; some elements were not recorded but were based on operator memory. For example, the time required to place the cannula was not measured precisely, and because different methods were used by different operators, certain steps were included for some operators and not for others. For the hemolysis evaluation, all patients received anticoagulation with nonfractionated heparin, and the circuits that were used had similar characteristics. However, because data regarding the circuits used and protocols for anticoagulation were not recorded, we cannot exclude that these variables may at least partly explain our findings. Similarly, apart from the damage observed after cannula placement, it is difficult to attribute certain complications that were encountered over the course of supportive care for the procedure. We were unable to measure the amount of vascular recirculation to evaluate the efficiency of the cannula. The manufacturer advertises this product as a solution to the problems encountered with multiple lines. To evaluate this ability, a prospective data collection protocol is needed.
Based on the entire dataset, we can make several recommendations that are important for the use of these cannulas. These recommendations must be confirmed by future studies. We propose the preferential use of a large-caliber (27 or 31-Fr.) cannula in adults to achieve the best flow rate, particularly in hypoxemic patients. Complications do not seem to increase with these sizes or whether an intensivist or surgeon places the cannula. We believe that it is possible to position the inferior end of the cannula under TTE control (subxyphoid view) before fixation. TEE use should be systematized in cases of decreased efficiency of support (or other events that suggest recirculation as a result of improper orientation of the ports toward the right ventricle) after TTE ensures the proper position of the distal end of the cannula in the inferior vena cava. Chest radiography enables daily monitoring of the cannula position, but TTE should be used regularly to control the position of the distal end of the cannula in the inferior vena cava. In cases of insufficient flow, we propose that the physician add a second line to the dual-lumen cannula rather than replacing it with a single-lumen catheter. Upon removal, we advise simple compression with possible cutaneous sutures. If hemostasis is not satisfactory, then we propose surgical management with jugular vein sutures. The available mobilization opportunities, ideally for patients with no supportive care, should be strongly promoted. Future studies should be conducted to evaluate the efficiency and safety of these recommendations.
The Avalon© bicaval cannula seems to be a good option for use in venovenous assistance. It has several advantages in terms of mobilization without altering the efficiency of supportive flow. These devices are highly efficient, notably in terms of supportive flow. The main complications are secondary displacement in the right ventricle. Percutaneous placement and removal are straightforward and can be performed by either a surgeon or an intensivist. Standardization of these procedures would prevent the complications described in the current study.
Dr. Edouard Sage, Service de Chirurgie Thoracique, Hôpital Foch, Suresnes; Dr. Vladimir Saplacan, Pr Massimo Massetti, Service de Chirurgie Cardiaque CHU de Caen; Dr. Anne Médard, Département d’Anesthésie-Réanimation, CHU Gabriel-Montpied Clermont-Ferrand; Dr. Nawwar Al Attar, Service de Chirurgie Cardiaque et Vasculaire, CHU Bichat-Claude Bernard; Dr. Erwan Flecher, Service de Chirurgie Cardio-vasculaire, Hôpital Pontchaillou CHU de Rennes; Dr. Eustache Marie-Line, Service de Réanimation Polyvalente, Centre Hospitalier Bretagne Atlantique Vannes; Dr. Elie Zogheib, Département d’Anesthésie, Hôpital Sud CHU d’Amiens Picardie; Dr. Jacob Eliet, Département d’anesthésie Réanimation Cœur Poumon, Hôpital Arnaud de Villeneuve CHU Montpellier; Dr. Karim Bouabdallah, Service de Réanimation Cardiologique, Institut Mutualiste de Monssouris Paris; Dr. Elmi Messai, Service de Réanimation, Centre Hospitalier de Cholet; Dr. Jean-Pierre Quenot, Service de Réanimation Médicale, CHU de Dijon; Dr. Hadrien Rozé, Département d’Anesthésie Réanimation, Hôpital Haut Lévêque CHU Bordeaux; Dr. Damien Bedague, Département d’Anesthésie Réanimation, CHU Grenoble; Dr. Thibaut Petroni, Pr Alain Combes, Service de Réanimation Médicale, CHU la Pitié Salpêtrière Paris; Dr. Philippe Goutorbe, Service de Réanimation, Hôpital d’Instruction des Armée Sainte Anne Toulon; Dr. Antoine Poidevin, Service de Réanimation Médicale, Centre Hospitalier de Mulhouse; and Dr. Jean-Marie Quintard, Service de Réanimation, Centre Hospitalier de Perpignan.
1. Bersten AD, Edibam C, Hunt T, Moran JAustralian and New Zealand Intensive Care Society Clinical Trials Group. . Incidence and mortality of acute lung injury and the acute respiratory distress syndrome in three Australian States. Am J Respir Crit Care Med. 2002;165:443–448
2. Brun-Buisson C, Minelli C, Bertolini G, et al. Epidemiology and outcome of acute lung injury in European intensive care units. Results from the ALIVE study. Intensive Care Med. 2004;30:51–61
3. Rubenfeld GD, Herridge MS. Epidemiology and outcomes of acute lung injury. Chest. 2007;131:554–562
4. Vasilyev S, Schaap RN, Mortensen JD. Hospital survival rates of patients with acute respiratory failure in modern respiratory intensive care units. An international, multicenter, prospective survey. Chest. 1995;107:1083–1088
5. Vieillard-Baron A, Girou E, Valente E, et al. Predictors of mortality in acute respiratory distress syndrome. Focus On the role of right heart catheterization. Am J Respir Crit Care Med. 2000;161:1597–1601
6. Herridge MS, Tansey CM, Matté A, et al.: Canadian Critical Care Trials Group. Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med. 2011;364:1293–1304
7. Zapol WM, Snider MT, Hill JD, et al. Extracorporeal membrane oxygenation in severe acute respiratory failure. A randomized prospective study. JAMA. 1979;242:2193–2196
8. Peek GJ, Mugford M, Tiruvoipati R, et al.: 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. 2009;374:1351–1363
9. Davies A, Jones D, Bailey M, et al. Extracorporeal membrane oxygenation for 2009 Influenza A (H1N1) acute respiratory distress syndrome. JAMA. 2009;302:1888–1895
10. Garcia JP, Iacono A, Kon ZN, Griffith BP. Ambulatory extracorporeal membrane oxygenation: A new approach for bridge-to-lung transplantation. J Thorac Cardiovasc Surg. 2010;139:e137–e139
11. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double lumen cannula-toward a percutaneous and ambulatory paracorporeal artificial lung. ASAIO J. 2008;54:606–611
12. Bermudez CA, Rocha RV, Sappington PL, Toyoda Y, Murray HN, Boujoukos AJ. Initial experience with single cannulation for venovenous extracorporeal oxygenation in adults. Ann Thorac Surg. 2010;90:991–995
13. Marasco SF, Lukas G, McDonald M, McMillan J, Ihle B. Review of ECMO (extra corporeal membrane oxygenation) support in critically ill adult patients. Heart Lung Circ. 2008;17 Suppl 4:S41–S47
14. Tabak B, Elliott CL, Mahnke CB, Tanaka LY, Ogino MT. Transthoracic echocardiography visualization of bicaval dual lumen catheters for veno-venous extracorporeal membrane oxygenation. J Clin Ultrasound. 2012;40:183–186
15. Javidfar J, Brodie D, Wang D, et al. Use of bicaval dual-lumen catheter for adult venovenous extracorporeal membrane oxygenation. Ann Thorac Surg. 2011;91:1763–8 discussion 1769
16. Cornejo R, Tobar E, Díaz G, et al. Systematic approach for severe respiratory failure due to novel A (H1N1) influenza. Minerva Anestesiol. 2011;77:510–521