Venovenous extracorporeal membrane oxygenation (ECMO) has the ability to support gas exchange in severe respiratory failure by directly oxygenating and removing carbon dioxide from the blood. Deoxygenated venous blood is withdrawn from a central vein, pumped through a gas exchange device, referred to as a membrane oxygenator, and reinfused into a central vein (Figure 1).1 The reinfused, oxygenated blood then enters the right atrium and passes through the patient’s native pulmonary and systemic circulation. The effectiveness of ECMO in supporting oxygenation is dependent on multiple factors, including the amount of blood flow through the device, the patient’s cardiac output and metabolic demand, the fraction of oxygen in the sweep gas, the diffusion properties and surface area of the oxygenator membrane, and the amount of recirculation present within the circuit.2–4 Recirculation occurs when reinfused oxygenated blood is withdrawn through the drainage cannula without passing through the systemic circulation (Figure 2), a phenomenon that does not exist in venoarterial or arteriovenous ECMO circuits.5,6 In femoral venoarterial ECMO, when there is residual native cardiac output and impaired native gas exchange, differential circulation may occur. Poorly oxygenated blood will supply the upper body, whereas the lower body will be perfused by well-oxygenated blood. One consequence of this differential circulation is differential venous oxygenation: blood with a higher venous oxygen saturation returning to the inferior vena cava (IVC) relative to the superior vena cava (SVC). The well-saturated blood returning to the IVC may be taken up by the drainage cannula, a phenomenon which may be interpreted as a form of recirculation. However, because the blood has already passed through the systemic circulation before reuptake by the circuit, this would not constitute recirculation in the classical sense.
It is important to emphasize that because recirculated blood does not contribute to systemic oxygenation, it decreases the efficiency with which ECMO provides oxygenation. Whether the amount of recirculation has clinical significance depends on the degree to which the patient is dependent on ECMO for oxygenation. This review will discuss the factors that contribute to recirculation of oxygenated blood, methods used to quantify recirculation, and interventions that may reduce the amount of recirculation in venovenous ECMO.
We conducted a literature search of all published manuscripts in MEDLINE. Medical Subject Headings (MeSH) terms used to identify relevant articles included ECMO, recirculation, SvO2, oxygenation, configuration, and cannulation strategies. Both human and animal studies were included. This review was designed as a narrative review of the existing literature, with recommendation for the management of recirculation based on published data and our own experience.
Factors Affecting Recirculation
Cannulation Configuration and Positioning
The traditional two-site, femoral-to-internal jugular venovenous ECMO configuration, with cannulae outflow and inflow ports in the IVC and SVC, respectively, lends itself to recirculation because the reinfusion jet is directed toward the drainage port (Figure 2). The proximity of the reinfusion and drainage ports will, therefore, have a direct impact on the amount of recirculation, with a higher percentage of recirculated blood flow when the two ports are in closer proximity. This factor may be particularly relevant in the neonatal and pediatric population where the SVC and IVC are in closer proximity. Patient position, such as movement from a supine to a seated position or rotation of the head and neck, may likewise affect the orientation of the cannulae, affecting the amount of recirculation.7 Published data supporting these finding are lacking, and this remains an area that would benefit from further research.
Pump Speed, Cannula Size, and Extracorporeal Blood Flow
Multiple studies have shown a direct correlation among pump speed, extracorporeal blood flow rates, and recirculation, although the recirculation rate varies greatly by cannula type, size, and position.8–10 Sreenan et al.10 demonstrated recirculation rates ranging from 30% to 70% with pump flows of 200–500 ml/min in a venovenous circuit via a unicaval dual-lumen cannula. In a study comparing different methods of calculating recirculation, consistent relationships with high correlation coefficients were identified for each of the ultrasound dilution, SvO2, and central venous line (CVL) methods (% recirculation = 1.88 + 0.23 × ECMO flow, r = 0.99; 1.42 + 0.32 × ECMO flow, r = 0.99; and 6.66 + 0.36 ×ECMO flow, r = 0.96, respectively).8 The relationship between ECMO blood flow rates and recirculation may, in part, be explained by an increase in negative venous pressures within the venous drainage limb that occurs at higher flow rates.11 Relatively larger drainage cannulae may allow for comparable blood flow rates at lower pump speeds with less negative venous pressure, potentially mitigating the amount of recirculation (Figures 3 and 4).
Intrathoracic, Intracardiac, and Intra-Abdominal Pressures
Pressures within the thoracic cavity, the heart, and the abdomen will have a variable effect on the amount of recirculation. Increases in tidal volume and decreases in chest wall compliance have been shown to increase central venous, pericardial, and pleural pressures and to decrease right and left heart output.12,13 In cases of marked elevations in intrathoracic and cardiac end-diastolic pressures, such as those caused by pneumothoraces or pericardial tamponade, venous return to the heart may be impeded even further, with both native and reinfused blood flow preferentially directed toward the drainage cannula. In extreme circumstances, ECMO blood flow may cease due to severe impairment of venous return.14 Excessively high intra-abdominal pressures may lead to compression of the IVC, along with elevations in central venous pressures,15 potentially limiting drainage and causing interruptions in extracorporeal blood flow. These effects may decrease the absolute amount of recirculation, although the effect on percentage recirculation will be less predictable. However, the decreased flow into the circuit itself would lessen the contribution of the device to systemic oxygenation.
Direction of Extracorporeal Blood Flow
The direction of drainage and reinfusion also has a significant impact on the amount of recirculation. Rich et al.16 studied the impact of extracorporeal blood flow direction on ECMO efficiency in 10 human subjects. Each subject was cannulated in the femoral and internal jugular veins with a bridge configuration that allowed for immediate alternation between femoral drainage and atrial reinfusion (femoro-atrial) and atrial drainage and femoral reinfusion (atrio-femoral). Femoro-atrial flow achieved a higher maximum blood flow (55.6 vs. 51.1 ml/kg/min, p = 0.04) and mixed venous oxygen saturation (89.9% vs. 83.2%, p = 0.006) than atrio-femoral flow. In addition, the amount of flow needed to achieve a given pulmonary arterial mixed venous oxygen saturation was lower with femoro-atrial flow (37.0 vs. 46.4 ml/kg/min, p = 0.04), suggesting that there is less recirculation in the femoro-atrial direction.
Estimating the Amount of Recirculation
The fraction of recirculation within the ECMO circuit can be quantified with the following equation:
where SpreO2 is the saturation of blood entering the oxygenator, SpostO2 is the saturation of blood leaving the oxygenator, and SvO2 is defined here as the saturation of venous blood returning to the vena cavae just before being drained by the ECMO circuit.
SpreO2 and SpostO2 can be measured either with noninvasive oximetry or by blood gas analyses from blood entering and exiting the oxygenator. If SpreO2 is equal to SvO2, then there is no recirculation in the circuit. If, however, SpreO2 is equal to SpostO2, then there is 100% recirculation.
This approach to quantifying recirculation is limited because SvO2 is difficult to measure. SvO2 is not equivalent to the traditional mixed venous saturation that may be obtained by sampling blood from the pulmonary artery because the ECMO circuit reinfuses oxygenated blood that then passes through the native circulation. Methods that have been proposed to estimate the SvO2 include the CVL method and the SvO2 method.3,8
The CVL method consists of measuring the venous saturation of blood from either the SVC or the IVC via a central venous catheter.3,8,18 Although this measurement can easily be obtained, the value will not accurately reflect the mixed venous saturation of blood from both the SVC and the IVC.
The SvO2 method involves turning off the sweep gas flow and using the ventilator to achieve an arterial oxygen saturation equivalent to what was achieved with ECMO support.8 The saturation of blood drained from the patient by the venous drainage cannula (as measured by noninvasive oximetry in the ECMO circuit) is then considered to represent SvO2.3,8 This method, which is more accurate than the CVL method in determining the true mixed venous saturation, is difficult—and potentially dangerous—to implement clinically in patients who are dependent on the ECMO circuit for gas exchange and would not tolerate an interruption in sweep gas flow.
A variation in the SvO2 method has been described both in animal models19 and in neonates9 receiving venovenous ECMO via dual-lumen cannulae, with recirculation assessed in the neonate study at the end of each ECMO run when the subjects could tolerate suspension of sweep gas flow.9 SvO2 was first measured with the sweep gas flow turned off (“SvO2”) and again with sweep gas set at 100% oxygen (“SQO2”). Measurements were then repeated at various blood flow rates. The amount of recirculation was calculated as:
In the neonatal study, recirculation increased with increasing amounts of blood flow, with rates as high as 65% at 500 ml/min of extracorporeal blood flow.
Ultrasound dilution is an alternative method of quantifying recirculation, one that is not dependent on measuring SvO2. First used in the context of arteriovenous fistula flow in dialysis in the 1990s,20,21 this method uses ultrasound wave velocity properties and dilution of blood within the extracorporeal circuit to noninvasively estimate the amount of recirculation.18 Saline is injected into the reinfusion limb of the circuit, and an ultrasound dilution sensor on this limb measures the velocity of ultrasound waves passing through that diluted portion of blood (Figure 5). A second sensor measures the velocity of ultrasound waves passing through blood entering the drainage limb. The ratio of these two velocities represents the percentage of recirculation in the circuit. If no dilution is detected in the drainage limb, then there is no recirculation present, whereas if the same amount of dilution is detected in both the drainage and the reinfusion limbs, then there is 100% recirculation.
The ultrasound dilution method has been compared with the CVL method in a swine model, with comparable amounts of recirculation detected by both methods at various blood flow rates (ultrasound vs. CVL: 0% vs. 2% at ECMO blood flow of 210 ml/min, p = nonsignificant [NS]; 8% vs. 8.6% at 410 ml/min, p = NS; 17% vs. 17.3% at 610 ml/min, p = NS).18 When ultrasound dilution has been compared with both the CVL and the SvO2 methods at progressively increasing blood flow rates in a sheep model, ultrasound dilution recirculation rates were more similar to those detected by the SvO2 method than the CVL method.8 Regardless of the method used, increasing blood flow consistently in a femoro-atrial cannulation approach resulted in an increasing amount of recirculation. It is important to note that none of these techniques is easily applied at the bedside. The precise determination of percentage recirculation is therefore difficult to ascertain clinically.
The ability to calculate SvO2 has been evaluated in vitro by interfacing a mixing chamber with a venovenous ECMO circuit, set to varying known (“actual”) SvO2 values, while maintaining a constant recirculation rate (determined by ultrasound dilution).22,23 Calculated SvO2 was derived from an equation incorporating SpreO2 and SpostO2, and percentage recirculation. Initial assessments found a close correlation between actual and calculated values when the actual SvO2 was low (31.8 ± 13.95% vs. 37.0 ± 6.7%, p = NS), but correlation was poor at higher actual SvO2 values (61.7 ± 1.5% vs. 72.3 ± 1.8%, p < 0.05, high: 84.4 ± 0.9% vs. 91.2 ± 1.1%, p < 0.05).22 Subsequent adjustments accounting for the partial pressure of oxygen improved the correlation between actual and calculated SvO2.23
Thermodilution has likewise been described as a noninvasive way of quantifying recirculation.10 In a rabbit model of venovenous ECMO using a dual-lumen cannula, ice cold saline was injected into the reinfusion limb, and temperature changes were measured in the drainage limb with the addition of a thermistor-tipped catheter in the limb. Percentage recirculation was estimated as 19 + 0.1 × pump flow, with a correlation coefficient (r) of 0.9, demonstrating a consistent relationship between recirculation and increases in extracorporeal flow.10 Although the absolute relationship may vary by cannula type, size, and position, this method may be useful to trend changes in recirculation within a given circuit over time. Of note, this method has not been validated in human subjects.
In our experience, the most practical method of estimating clinically relevant recirculation is trending the SpreO2 and peripheral arterial oxygen saturation (SaO2).24,25 Increasing SpreO2 in the setting of decreasing SaO2 is indicative of clinically relevant recirculation and is most apparent when SpreO2 exceeds SaO2, a scenario that cannot otherwise be explained in the absence of recirculation. Any attempt to reduce recirculation, if successful, should result in a decrease in SpreO2 with an increase in SaO2.
Interventions to Reduce Recirculation
Increasing Distance Between Cannulae
As previously discussed, the proximity of drainage and reinfusion ports will have an effect on the amount of recirculation. There is no standard distance that should be maintained between ports. If clinically significant recirculation is suspected, then one strategy is to increase the distance between drainage and reinfusion cannulae by withdrawal of either cannula (Figure 6). Based on previously published data by Rich et al.,16 along with our own experience, we recommend that the tip of the femoral drainage cannula be positioned near the atrial-IVC junction with the side ports located within the hepatic IVC, as this portion of the IVC is least collapsible due to the surrounding hepatic parenchyma and least likely to impair drainage by occlusion of the ports when under negative pressure. This may limit how far the drainage cannula may be withdrawn. Close attention should be paid to the SpreO2 and blood flow rates as the cannula is manipulated to assess for both changes in recirculation and stability of blood flow.
Additional Drainage Cannula
As previously discussed, recirculation is, in part, affected by extracorporeal blood flow rates, in proportion to pump speed and drainage cannula size. The addition of a second drainage cannula may allow for comparable blood flow rates to be achieved at lower pump speeds, resulting in the generation of less negative venous pressure.26
Manipulation of the Reinfusion Cannula
Alterations to the shape or orientation of the reinfusion cannula may decrease the amount of recirculation. Lin et al.27 described using sutures to create a curvature in the cannula such that the reinfusion jet was directed toward the tricuspid valve. The resulting improvement in arterial saturation was attributed to a reduction in recirculation. An alternative configuration that has been proposed involves the combination of a curved reinfusion cannula directed toward the tricuspid valve and placement of the venous drainage cannula with its distal tip in the SVC-right atrial junction (Figure 7).28 This cannulation strategy, termed the x-configuration, was studied in a small, randomized controlled trial of 30 patients receiving venovenous ECMO for severe ARDS. The x-configuration group (n = 16) had a statistically significant reduction in recirculation, measured by the CVL method, compared with the traditional two-site configuration (n = 14) (5.3% vs. 29.4%, p < 0.0001). The x-configuration was associated with an ability to reduce the amount of invasive mechanical ventilatory support while simultaneously improving systemic oxygenation.
Veno-Right Ventricular Cannulation
Yet, another modification to the traditional two-site configuration, studied in a canine model, consists of drainage from the IVC and reinfusion through a cannula placed through the tricuspid valve and into the right ventricle.29 This strategy, referred to as veno-right ventricular ECMO, resulted in significantly less recirculation at 4 L/min of blood flow (8.4% vs. 37.9%, p = 0.008) and less of an increase in recirculation for every 1 L change in blood flow (2.9% vs. 11.1%, p < 0.0001). However, placement of the reinfusion cannula into the right ventricle was associated with endocardial damage and arrhythmias, which limits its clinical application.
Use of a Bicaval Dual-Lumen Cannula
The bicaval, dual-lumen cannula is one of the most significant advancements in cannula design impacting recirculation. The cannula has undergone several modifications since its original design in the 1980s.24,30–33 Initially, the cannula was designed to be unicaval, with its tip terminating in the right atrium, and drainage and reinfusion ports were in relatively close proximity.31,32 In an animal model of ECMO for severe respiratory failure, venovenous ECMO with this cannula design was able to provide full respiratory support.31 However, the venovenous configuration, compared with a venoarterial configuration, required higher circuit blood flow rates (124 vs. 101 ml/kg/min) and achieved lower systemic arterial oxygenation (50 vs. 247 mm Hg). These differences in blood flow rates and oxygen tension were attributed to the presence of recirculation, which does not exist in venoarterial ECMO. Subsequent adjustments in cannula design included an increase in the distance between drainage and reinfusion ports.24 When compared with the original design, the cannula with the newer port positions resulted in higher pulmonary arterial and cerebral saturations and significantly lower mixed venous saturations at varying extracorporeal blood flow rates, suggesting a reduction in recirculation with the newer cannula design.24 More recent modifications to the dual-lumen cannula have included a bicaval design, with the distal tip of the cannula situated in the IVC. Drainage ports are located in both the SVC and the IVC, and the reinfusion jet is intended to be directed toward the tricuspid valve (Figure 8).34,35 This cannula has the advantage of using a single vascular access point for venovenous extracorporeal support. When properly positioned, which usually requires echocardiographic or fluoroscopic guidance,36 recirculation may be as low as 2%,34 which is significantly lower than recirculation rates reported with older cannula designs.9 However, malposition of the cannula can significantly increase the degree of recirculation and compromise the efficiency of the circuit, with recirculation rates reported to be as high as 50% in a sheep model.34 We recommend that any adjustment of the cannula after its initial insertion be done under echocardiographic guidance to ensure appropriate orientation of the reinfusion jet. The combination of echocardiography and ultrasound dilution may further optimize cannula positioning.37
Recirculation compromises the ability of ECMO to support systemic oxygenation; however, the precise amount of recirculation can be difficult to quantify. Efforts to decrease recirculation are helpful in maximizing oxygen delivery, although the need to do so should primarily be dictated by whether the patient’s systemic oxygen requirements are being met. Advances in extracorporeal cannulation strategies, particularly the development of the dual-lumen cannula, have helped to minimize the amount of recirculation in venovenous ECMO.
1. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation
for ARDS in adults. N Engl J Med. 2011;365:1905–1914
2. Agerstrand CL, Bacchetta MD, Brodie D. ECMO
for adult respiratory failure: Current use and evolving applications. ASAIO J. 2014;60:255–262
3. Cornish D, Clark RZwischenberger J, Bartlett R. Principles and practice of venovenous extracorporeal membrane oxygenation
. In ECMO
, Extracorporeal Cardiopulmonary Support in Critical Care. 1995 Ann Arbor, MI ELSO:87–107
4. 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;39:838–846
5. Abrams D, Combes A, Brodie D. Extracorporeal membrane oxygenation
in cardiopulmonary disease in adults. J Am Coll Cardiol. 2014;63(25 pt A):2769–2778
6. Gail M. Annich, William R. Lynch, Graeme MacLaren, et al. Extracorporeal Cardiopulmonary Support in Critical Care. 20124th ed. Ann Arbor, MI Extracorporeal Life Support Organization
7. Clements D, Primmer J, Ryman P, Marr B, Searles B, Darling E. Measurements of recirculation
during neonatal veno-venous extracorporeal membrane oxygenation
: Clinical application of the ultrasound dilution technique. J Extra Corpor Technol. 2008;40:184–187
8. van Heijst AF, van der Staak FH, de Haan AF, et al. Recirculation
in double lumen catheter veno-venous extracorporeal membrane oxygenation
measured by an ultrasound dilution technique. ASAIO J. 2001;47:372–376
9. Anderson HL 3rd, Otsu T, Chapman RA, Barlett RH. Venovenous extracorporeal life support in neonates using a double lumen catheter. ASAIO Trans. 1989;35:650–653
10. Sreenan C, Osiovich H, Cheung PY, Lemke RP. Quantification of recirculation
by thermodilution during venovenous extracorporeal membrane oxygenation
. J Pediatr Surg. 2000;35:1411–1414
11. Park M, Costa E, Maciel A, et al. Effect of flow rate and temperature on transmembrane blood pressure drop in an extracorporeal artificial lung. Perfusion. 2014;29:517–525
12. Lansdorp B, Hofhuizen C, van Lavieren M, et al. Mechanical ventilation-induced intrathoracic pressure distribution and heart-lung interactions*. Crit Care Med. 2014;42:1983–1990
13. Mesquida J, Kim HK, Pinsky MR. Effect of tidal volume, intrathoracic pressure, and cardiac contractility on variations in pulse pressure, stroke volume, and intrathoracic blood volume. Intensive Care Med. 2011;37:1672–1679
14. Zwischenberger JB, Cilley RE, Hirschl RB, Heiss KF, Conti VR, Bartlett RH. Life-threatening intrathoracic complications during treatment with extracorporeal membrane oxygenation
. J Pediatr Surg. 1988;23:599–604
15. Rosenthal RJ, Friedman RL, Kahn AM, et al. Reasons for intracranial hypertension and hemodynamic instability during acute elevations of intra-abdominal pressure: Observations in a large animal model. J Gastrointest Surg. 1998;2:415–425
16. Rich PB, Awad SS, Crotti S, Hirschl RB, Bartlett RH, Schreiner RJ. A prospective comparison of atrio-femoral and femoro-atrial flow in adult venovenous extracorporeal life support. J Thorac Cardiovasc Surg. 1998;116:628–632
17. Fortenberry JD, Pettignano R, Dykes FVan Meurs K, Lally D, Peek G, Zwischenberger J. Principles and practice of venovenous ECMO
. In ECMO
Extracorporeal Cardiopulmonary Support in Critical Care. 20053rd ed Ann Arbor, MI Extracorporeal Life Support Organization:94
18. Darling EM, Crowell T, Searles BE. Use of dilutional ultrasound monitoring to detect changes in recirculation
during venovenous extracorporeal membrane oxygenation
in swine. ASAIO J. 2006;52:522–524
19. Okamoto T, Ichinose K, Tanimoto H, et al. Preliminary experiment with a newly developed double balloon, double lumen catheter for extracorporeal life support vascular access. ASAIO J. 2003;49:583–588
20. Krivitski NM. Theory and validation of access flow measurement by dilution technique during hemodialysis. Kidney Int. 1995;48:244–250
21. Depner TA, Krivitski NM, MacGibbon D. Hemodialysis access recirculation
measured by ultrasound dilution. ASAIO J. 1995;41:M749–M753
22. Walker J, Primmer J, Searles BE, Darling EM. The potential of accurate SvO2 monitoring during venovenous extracorporeal membrane oxygenation
: An in vitro
model using ultrasound dilution. Perfusion. 2007;22:239–244
23. Walker JL, Gelfond J, Zarzabal LA, Darling E. Calculating mixed venous saturation during veno-venous extracorporeal membrane oxygenation
. Perfusion. 2009;24:333–339
24. Raïs-Bahrami K, Rivera O, Mikesell GT, Seale WR, Sell JE, Short BL. Improved oxygenation with reduced recirculation
during venovenous extracorporeal membrane oxygenation
: Evaluation of a test catheter. Crit Care Med. 1995;23:1722–1725
25. Rais-Bahrami K, Walton DM, Sell JE, Rivera O, Mikesell GT, Short BL. Improved oxygenation with reduced recirculation
during venovenous ECMO
: Comparison of two catheters. Perfusion. 2002;17:415–419
26. Ichiba S, Peek GJ, Sosnowski AW, Brennan KJ, Firmin RK. Modifying a venovenous extracorporeal membrane oxygenation
circuit to reduce recirculation
. Ann Thorac Surg. 2000;69:298–299
27. Lin TY, Horng FM, Chiu KM, Chu SH, Shieh JS. A simple modification of inflow cannula to reduce recirculation
of venovenous extracorporeal membrane oxygenation
. J Thorac Cardiovasc Surg. 2009;138:503–506
28. Bonacchi M, Harmelin G, Peris A, Sani G. A novel strategy to improve systemic oxygenation in venovenous extracorporeal membrane oxygenation
: The “χ-configuration.” J Thorac Cardiovasc Surg. 2011;142:1197–1204
29. Lindstrom SJ, Mennen MT, Rosenfeldt FL, Salmonsen RF. Veno-right ventricular cannulation
in extracorporeal membrane oxygenation
. Perfusion. 2012;27:464–469
30. Zwischenberger JB, Toomasian JM, Drake K, Andrews AF, Kolobow T, Bartlett RH. Total respiratory support with single cannula venovenous ECMO
: Double lumen continuous flow vs. single lumen tidal flow. Trans Am Soc Artif Intern Organs. 1985;31:610–615
31. Andrews AF, Zwischenberger JB, Cilley RE, Drake KL. Venovenous extracorporeal membrane oxygenation
) using a double-lumen cannula. Artif Organs. 1987;11:265–268
32. Otsu T, Merz SI, Hultquist KA, et al. Laboratory evaluation of a double lumen catheter for venovenous neonatal ECMO
. ASAIO Trans. 1989;35:647–650
33. Durandy Y, Chevalier JY, Lecompte Y. Single-cannula venovenous bypass for respiratory membrane lung support. J Thorac Cardiovasc Surg. 1990;99:404–409
34. 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
35. Javidfar J, Brodie D, Wang D, et al. Use of bicaval dual-lumen catheter for adult venovenous extracorporeal membrane oxygenation
. Ann Thorac Surg. 2011;91:1763–1768; discussion 1769
36. Javidfar J, Wang D, Zwischenberger JB, et al. Insertion of bicaval dual lumen extracorporeal membrane oxygenation
catheter with image guidance. ASAIO J. 2011;57:203–205
37. Körver EP, Ganushchak YM, Simons AP, Donker DW, Maessen JG, Weerwind PW. Quantification of recirculation
as an adjuvant to transthoracic echocardiography for optimization of dual-lumen extracorporeal life support. Intensive Care Med. 2012;38:906–909