Secondary Logo

Journal Logo

Respiratory Support

In Vitro Performance Testing of a Pediatric Oxygenator With an Integrated Pulsatile Pump

Borchardt, Ralf; Schlanstein, Peter; Mager, Ilona; Arens, Jutta; Schmitz-Rode, Thomas; Steinseifer, Ulrich

Author Information
doi: 10.1097/MAT.0b013e318251dc70
  • Free


Since the early 1970s extracorporeal life support (ECLS) has been reported as an efficacious therapy for lung diseases.1–3 extracorporeal life support uses an oxygenator to support the patients’ lungs and maintain sufficient gas exchange. Most modern ECLS circuits incorporate membrane oxygenators,4 hence the reason why extracorporeal membrane oxygenation (ECMO) has become a common term. Compared with mechanical ventilation the treatment with an extracorporeal circuit is gentler for the patients’ lungs in terms of pulmonary edema5 but is a more complex therapy, which is available only in specialized clinical centers.

Common indications to use ECMO as a treatment for respiratory failures include severe neonatal sepsis, meconium aspiration syndrome, congenital diaphragmatic hernia, and respiratory distress syndrome.2,6 Extracorporeal membrane oxygenation is also prescribed as cardiac support in specific conditions ranging from congenital cardiac defects and cardiomyopathy to use as bridge to heart transplant.6–8 Even though the success rate of ECMO treatment is approximately 75% depending on the underlying medical condition,9 the total number of treatment applications is still small. Reasons behind this moderate practice of ECMO therapy are complications such as bleeding and thrombosis associated with ECMO support.1,9 Most reported cases of ECMO treatment involve neonatal (<31 days of age) and pediatric (<18 years of age) patients. Until 2004, 66% of all ECMO patients were neonates with respiratory failure. The overall survival of these patients to discharge or transfer was 77%.6 Up to July 2008, 35,050 recorded ECMO cases were associated with patients under the age of 18 years.8 It has been assumed that the number of ECMO applications would increase if the therapy is optimized in terms of simplified usage and design adjustments to components that are developed specifically for this therapy.2

Currently, different approaches are followed to improve the devices of extracorporeal circuits. Often the major challenges with ECMO circuits are associated with minimizing priming volume and surface area of blood contacting surfaces.9–12 These are important issues, especially for neonatal applications because of their low body weight. One approach to minimizing priming volume and blood contacting surface area is to optimize the gas exchange efficiency of oxygenators. Shunt flows and recirculation areas should be avoided and an equalized flow pattern through the oxygenator fiber bundle should be generated.4,11,13

Another approach is to combine different components into one compact device. This permits flow path optimization of components together and saves priming volume introduced by connecting tubes. Currently, there are several commercial oxygenators with integrated heat exchangers available. There have also been efforts to integrate a pump into an oxygenator; different types of rotary pumps placed in the center of the oxygenator’s hollow-fiber bundle have been reported.9,11,14–18 This research group investigates a novel approach to pump integration by using collapsing and expanding silicone tubes placed in a fiber bundle in combination with active valves.19 In contrast to solutions with a rotary pump, this method enables uniform pump placement within the oxygenator. By placing the tubes systematically inside the bundle, blood mixing inside the oxygenator could be increased20 and the risk of shunt flows and recirculation areas could potentially be reduced. Additionally, the collapsing and expanding tubes generate pulsatile flow. Investigations from several research groups have been carried out, which reported the benefits of pulsatile flow in comparison to continuous flow in extracorporeal circuits.10,21,22 Most notably, in long-term applications pulsatile flow improves end-organ microcirculation.23 Nevertheless controversial discussions about benefits of pulsatile versus continuous flow are still ongoing. It has been found that continuous flow, which is 20% above the mean mass flow during pulsatile flow can achieve equivalent microcirculation.24 Additionally, mean flow for a continuous flow pump is not necessarily pathological if the device supports the metabolic demands of the tissue.25

This study presents the improved design of an oxygenator with integrated pumping silicone tubes, called ExMeTrA. The top criteria for improvements were minimization of the priming volume and maximization of gas exchange.


The intention of the ExMeTrA project is to develop an oxygenator with an integrated pulsatile pump that reduces the filling volume of current extracorporeal circuits. To investigate the gas-exchange efficiency as well as the pumping behavior an in vitro test series of five different ExMeTrA modules (Figure 1) was performed.

Figure 1
Figure 1:
A: Current version of the ExMeTrA oxygenator (flow direction from left to right); B: Bottom view and tube arrangement (flow direction from bottom to the top).


To manufacture the ExMeTrA oxygenators, 2 meters of a two-layer mat of OXYPLUS polymethylpentene fibers (Membrana GmbH, Wuppertal, Germany) in combination with 62 flexible silicone tubes (inner diameter = 2.0 mm; wall thickness = 0.15 mm; Raumedic GmbH, Helmbrechts, Germany) were wound upon an oval-shaped core. An equidistant placement of silicone tubes in critical regions of the fiber bundle was used to generate a specific tubular arrangement (Figure 1B). Stiff aluminum cores (inner diameter = 1.6 mm) were placed inside the silicone tubes to avoid collapse during coiling. The intended tube arrangement consistency was ensured within the single modules. Once wound the complete fiber bundle was placed inside the housing and the oxygenator fiber ends were sealed on a centrifuge using two-component biocompatible silicone (Elastosil RT 625; Wacker Chemie AG, Munich, Germany). This permits reuse of the oxygenator housing and core.

After sealing the oxygenator, the aluminum cores were removed from the silicone tubes and the oxygenator fiber ends were cut open. One side of the silicone tubes was sealed and the opposite side was connected to stiff polytetrafluoroethylene (PTFE) tubes. The PTFE tubes had dimension of outer diameter 1.79 mm, inner diameter of 1.19 mm, and were approximately 50 mm in length (Zeus, Orangeburg, SC). Subsequently, the oxygen and pressure chambers were placed on top of the oxygenator housing and the stiff PTFE tubes were sealed with silicone.

The major improvements in the manufacturing technique used to generate the modules as compared with the previous published version19 was the usage of a central core in the fiber bundle as well as the intended tubular arrangement. With regard to the priming volume, the use of a centrifuge to seal the fiber bundle achieved better consistency between different modules. Furthermore a much higher packing density of fibers inside the bundle was possible. Aluminum cores were used to avoid collapse of the silicone tubes during the coiling process. As the aluminum cores have a smaller diameter than the silicone tubes, the tubes adapt to the bundles shape (Figure 2). With these adjustments in manufacturing process the priming volume was reduced by approximately 50% from 44.3 ± 4.5 ml to 20 ± 2.2 ml. The fiber bundle’s height, however, remained sensitive to the amount of silicone used to seal the bundle, therefore there were still discrepancies between each of the modules regarding priming volume and surface area.

Figure 2
Figure 2:
Detailed view of sealed silicone tube. PMP, polymethylpentene.

General Configuration

The ExMeTrA oxygenator consists of three main chambers (Figure 3). Chamber III contains an oval-shaped fiber bundle consisting of 62 silicone tubes. The silicone tubes are connected to Chamber I by means of stiff PTFE tubes. An air pressure pulse, generated by an external pulsator device, is connected to Chamber I using a Luer taper (Value Plastics, Inc., Fort Collins, CO). The pulse frequency can be regulated between 20 and 140 bpm, and pressure differences up to 600 mmHg can be applied. The ratio of systolic to diastolic time can be varied from 0.2 to 0.8. The air pressure in Chamber I distributes through the stiff PTFE tubes into the silicone tubes. The change in pressure causes the thin-walled silicone tubes in Chamber III to collapse and expand. A triggering signal, provided by the external pulsator, opens and closes the active pinch-solenoid valves (Zimmer Automation, Rosengarten, Germany) that are placed externally at the tubes at blood inlet and blood outlet. Oxygen is connected to Chamber II by means of a Luer taper.

Figure 3
Figure 3:
Sectional view inside the ExMeTrA oxygenator.

One complete pulse cycle proceeds as follows: the outlet valve closes as the inlet valve opens. Negative pressure collapses the silicone tubes so that blood is actively drawn into Chamber III. Here the blood is enriched with oxygen, and carbon dioxide is removed. Afterward the inlet valve closes while the outlet valve opens. Positive pressure within Chamber I expands the silicone tubes, and the blood is pumped to the oxygenators’ blood outlet, and a new pulse starts. The pressure pulses as well as the valve closing signals are generated by a pulsator manufactured by BYTEC GmbH (Stolberg, Germany).

As each silicone tube actively pumps blood by collapsing and expanding, the blood is well distributed within the fiber bundle. In this manner the risk of shunt flows and recirculation areas within the oxygenator can be reduced. It can also be assumed that a well-distributed flow within the fiber bundle leads to a high gas-exchange efficiency.11,26

In Vitro Tests

This study contains five in vitro tests following the ISO 7199 for gas exchange measurements in oxygenators using porcine blood. Each test was performed with a newly manufactured module. Before starting the gas exchange measurements, porcine blood in a circuit within the reservoir, blood pump, and oxygenator was adjusted to venous values as mandated by ISO 7199 (oxyhemoglobin percentage: [65 ± 5]; hemoglobin: [12 ± 1] g/dl; base excess: [0 ± 5] mmol/l; partial pressure of carbon dioxide in blood, PCO2: [6.0 ± 0.7] kPa; temperature: [37 ± 1]°C). To adjust these values a Jostra roller pump (20-320 rpm, Jostra Inc., Austin, TX) was used in combination with a Hilite 7000 oxygenator (MEDOS Medizintechnik AG, Stolberg, Germany). Once the four liters inside the reservoir (MEDOS) were adjusted, the Hilite 7000 and roller pump were clamped. Subsequent to the clamping, gas exchange measurements started with venous blood from the reservoir. Different blood-volume flows through the ExMeTrA oxygenator were generated solely by the silicone tubes as a pump. Pure oxygen entered the system at a flow rate twice the blood flow rate through the oxygenator. Two blood samples were taken at the oxygenator’s inlet and four at the oxygenator’s outlet for each operating point (Figure 4). These samples were analyzed using the blood gas analyzer ABL 800 (Radiometer Medical ApS, Brønshøj, Denmark).

Figure 4
Figure 4:
In vitro test circuit.

Blood flow was measured using a flow-sensor (Transonic Systems Inc., Ithaca, NY) with integrated bubble alarm, set to highest sensitivity (alarm signal when the ultrasonic signal strength is less than 95% of the initial value). The bubble detector was used to detect possible air leakage through the oxygenator fibers caused by the collapsing tubes. After passing the flow sensor, the oxygenated blood was routed to the disposal.

Data Analysis

As the blood gas analyzer ABL 800 used to measure the blood samples cannot measure the oxygen fraction inside the blood, this value was calculated. For this purpose the following equation was used27:

corresponds to the blood volume flow, ΔPo2 equals the oxygen’s partial pressure difference from the oxygenator’s inlet to the outlet. As multiple samples were measured for each volume flow, arithmetic mean values were used in these equations. The hemoglobin concentration CHb as well as the oxygen saturation So2 were determined by the ABL. αo2blood however corresponds to oxygen’s overall solubility in blood as a sum of oxygen’s solubility in plasma αo2P and erythrocytes αo2rbc. Those can be calculated depending on the temperature T using the following equations28:

The carbon dioxide fraction however was calculated by the ABL 800 directly, using the following equation based on research of Siggaard-Andersen29:

corresponds to the blood volume flow again, and ΔPco2 equals the carbon dioxide’s partial pressure difference from the oxygenator’s inlet to the outlet. ΔctCO2(P) represents the carbon dioxide concentration difference between oxygenator’s inlet to outlet inside the blood’s plasma and can be calculated with the carbon dioxide’s partial pressure difference ΔPCO2 and the bicarbonate’s concentration difference ΔHCO3(P)2 inside the blood’s plasma:

pHEry and pKEry however consider the carbon dioxide’s dependence of oxygen saturation:


Blood flows up to 200 ml/min could be achieved by all five modules. One module had a maximum pumping capacity of 300 ml/min, two were able to pump up to 400 ml/min and one module could pump up to 500 ml/min. Pressures of ±70 mmHg inside the silicone tubes at frequencies of 21–25 bpm were sufficient for each module to achieve a flow rate of 100 ml/min. Only one module was able to pump up to 500 ml/min with pressures of +200/−100 mmHg inside the tubes and a frequency of 110 bpm. For the remaining modules these settings led to the aforementioned maximum flow rates. The mean gas exchange results of all five modules as well as achievable blood volume flows are illustrated in Figures 5 and 6.

Figure 5
Figure 5:
Averaged results of oxygen exchange for all five modules.(Note: Only one oxygenator achieved a flow rate of 500 ml/min)

Figures 5 and 6 illustrate the average gas exchange for all five modules. For blood flows of 100–500 ml/min the oxygen exchange rates were consistent and above 64mlo2/Lblood for each module and each blood flow. The corresponding oxygen saturations were ≥98%. The exchange rate for carbon dioxide was >62.5mlco2/Lblood for each blood flow with considerably higher values at low blood flows (Figure 6). The carbon dioxide partial pressure difference from the oxygenator inlet to outlet was 10.5–16.63 mmHg, depending on the blood flow rate.

Figure 6
Figure 6:
Averaged results of carbon dioxide exchange for all five modules.(Note: Only one oxygenator achieved a flow rate of 500 ml/min.)

During the test series a flow sensor with integrated bubble detector (Transonic Systems Inc.) was set to the highest sensitivity of 95% and used to measure the blood flow. Aside from priming the ExMeTrA oxygenator no bubbles were detected with this sensor, thus no air leakage through the fiber membranes caused by the collapsing silicone tubes was detected.


The findings from this study confirm previous investigations of development of a pulsatile pump by means of thin-walled silicone tubes inside an oxygenator as a promising technique. Improvements were made to the manufacturing of the modules comparing with former studies. The silicone tubular arrangement was deliberate thereby achieving a stronger cohesion of the modules. In this study, a symmetric and uniform silicone tube distribution was chosen. However, any user-defined tube configuration is possible. Additionally, the packing density of the fiber bundle was increased up to a total of 50%, which is comparable to commercially available oxygenators. Compared with the previous geometry, the priming volume could be reduced from 44 ml to 20 ± 2 ml, whereas the gas exchange capacity was increased. The usage of a central core inside the oxygenators bundle expedients manufacturing as it significantly simplifies the coiling of a bundle.

In comparison to commercially available oxygenators, the ExMeTrA oxygenators have less priming volume despite the integrated pump. Irrespective of the relatively high standard deviations during the testing, the gas exchange for oxygen as well as for carbon dioxide was sufficient for all blood flows. However, only one module produced the results for 500 ml/min therefore no final conclusions should be drawn out of this point alone. Moreover, gas exchange rates were comparable to those of commercial oxygenators for similar blood flow rates. This leads to the assumption that the gas exchange would be sufficient even for higher blood flows. Nevertheless the ExMeTrA oxygenators’ pumping capacity of 200–500 ml/min was lower than the expected calculated pumping volume of 692 ± 75 ml/min. Therefore, the next stage of development will be concerned with improving the pumping capacity without decreasing the gas exchange efficiency. As all five tested ExMeTrA oxygenators were able to pump 100 ml/min with comparable pump parameter settings (±70 mmHg, 21–25 bpm) the challenges in this next step seem to occur for higher frequencies. Hence, a potential design refinement to achieve higher blood flows would be modification of the inlet and outlet geometries. With an improved distribution of the blood before entering the fiber bundle it might be possible to pump with a higher efficiency even for higher frequencies. However, changes in the inlet and outlet geometries have the potential to increase the priming volume of the ExMeTrA oxygenator. Additionally, these changes could also cause hemolysis and thrombosis. Therefore, tests to investigate these effects will be completed with the future ExMeTrA designs. Long-term tests regarding the durability of the pumping silicone tubes as well as the durability of the silicone tubes in between the valves have to be performed in the future as well.

The effect of different pulsator settings on pumping performance and on gas exchange behavior of the oxygenator should be investigated as well. Future work will also investigate the possibility to use silicone tubes not only as a blood pump but also as a heat exchanger for oxygenators. By using tempered saline solution instead of air as a driving fluid to collapse and expand the silicone tubes it might be possible to temper the blood. The thermal conductivity of silicone is unpropitious compared with materials that are generally used in heat exchangers. But the flexible properties of silicone are essential to use the tubes as a pump. However, the thin-wall thickness of the silicone tubes might allow tempering the blood while pumping it through the oxygenator. The silicone tubes surface area that could be used as a heat exchanger is small in the current design (0.0132 ± 0.0013 m2). Considering the compact design of pump, oxygenator and heat exchanger this surface area might be sufficient to equalize the heat loss of a compact extracorporeal circuit. However, the conduction of the flow inside the silicone tubes has to be changed in this case. Thus, the tempered saline solution has to be pumped through the silicone tubes in a way that thermal energy is conducted to the device perpetually.


The feasibility of using silicone tubes inside an oxygenator as a pump to reduce priming volume of an extracorporeal circuit was demonstrated in this study. Improvements in the manufacturing process allowed for a significant reduction in the oxygenator’s priming volume and an increase in gas exchange rates as compared with previous studies. Nevertheless, further studies are necessary to enhance the pumping efficiency. A further benefit of silicone tubes inside the ExMeTrA oxygenator is the possibility to use them as a heat exchanger. By combining an oxygenator with a pulsatile pump as well as a heat exchanger, the filling volume of an extracorporeal circuit could be significantly reduced. As such a compact device could be placed close to the patient, long tubing is not necessary. This might lead to a more gentle therapy and could simplify ECMO applications in the future.


This work was performed in the course of a research and development project with the Dritte Patentportfolio Beteiligungsgesellschaft mbH & Co. KG, Schönefeld, Germany, who is also the owner of the underlying international patent.


1. Lewandowski K. Extracorporeal membrane oxygenation for severe acute respiratory failure. Crit Care. 2000;4:156–168
2. Bartlett RH. Extracorporeal life support: history and new directions. Semin Perinatol. 2005;29:2–7
3. Palanzo D, Qiu F, Baer L, Clark JB, Myers JL, Undar A. Evolution of the extracorporeal life support circuitry. Artif Organs. 2010;34:869–873
4. Haworth WS. The development of the modern oxygenator. Ann Thorac Surg. 2003;76:S2216–S2219
5. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am J Respir Crit Care Med. 1998;157:294–323
6. Conrad SA, Rycus PT, Dalton H. Extracorporeal Life Support Registry Report 2004. ASAIO J. 2005;51:4–10
7. Jeewa A, Manlhiot C, McCrindle BW, Van Arsdell G, Humpl T, Dipchand AI. Outcomes with ventricular assist device versus extracorporeal membrane oxygenation as a bridge to pediatric heart transplantation. Artif Organs. 2010;34:1087–1091
8. Haines NM, Rycus PT, Zwischenberger JB, Bartlett RH, Undar A. Extracorporeal Life Support Registry Report 2008: neonatal and pediatric cardiac cases. ASAIO J. 2009;55:111–116
9. Pantalos GM, Horrell T, Merkley T, et al. In vitro: characterization and performance testing of the ension pediatric cardiopulmonary assist system. ASAIO J. 2009;55:282–286
10. Tayama E, Niimi Y, Takami Y, et al. Effects of pulsatile flow on gas transfer of membrane oxygenator: MENOX EL-4000 and Gyro C1-E3 pulsatile mode. Artif Organs. 1997;21:1127–1132
11. Zhang J, Taskin ME, Koert A, et al. Computational design and in vitro characterization of an integrated maglev pump-oxygenator. Artif Organs. 2009;33:805–817
12. Schnoering H, Arens J, Terrada E, et al. A newly developed miniaturized heart-lung machine–expression of inflammation in a small animal model. Artif Organs. 2010;34:911–917
13. Bhavsar SS, Schmitz-Rode T, Steinseifer U. Numerical modeling of anisotropic fiber bundle behavior in oxygenators. Artif Organs. 2011;35:1095–1102
14. Tatsumi E. Artificial lungs: current state and trends of clinical use and research and development. J Artif Organs. 2007;10:1–5
15. Cattaneo G, Strauss A, Reul H. Compact intra- and extracorporeal oxygenator developments. Perfusion. 2004;19:251–255
16. Arens J, Schnöring H, Reisch F, Vázquez-Jiménez JF, Schmitz-Rode T, Steinseifer U. Development of a miniaturized heart-lung machine for neonates with congenital heart defect. ASAIO J. 2008;54:509–513
17. Zhang T, Cheng G, Koert A, et al. Functional and biocompatibility performances of an integrated Maglev pump-oxygenator. Artif Organs. 2009;33:36–45
18. Arens J, Schnoering H, Pfennig M, et al. The Aachen MiniHLM–a miniaturized heart-lung machine for neonates with an integrated rotary blood pump. Artif Organs. 2010;34:707–713
19. Borchardt R, Schlanstein P, Arens J, et al. Description of a flow optimized oxygenator with integrated pulsatile pump. Artif Organs. 2010;34:904–910
20. Eash HJ, Budilarto SG, Hattler BG, Federspiel WJ. Investigating the effects of random balloon pulsation on gas exchange in a respiratory assist catheter. ASAIO J. 2006;52:192–195
21. Agati S, Mignosa C, Ciccarello G, Dario S, Undar A. Pulsatile ECMO in neonates and infants: first European clinical experience with a new device. ASAIO J. 2005;51:508–512
22. Ji B, Undar A. An evaluation of the benefits of pulsatile versus nonpulsatile perfusion during cardiopulmonary bypass procedures in pediatric and adult cardiac patients. ASAIO J. 2006;52:357–361
23. Orime Y, Shiono M, Nakata K, et al. The role of pulsatility in end-organ microcirculation after cardiogenic shock. ASAIO J. 1996;42:M724–M729
24. Valdés F, Takatani S, Jacobs GB, et al. Comparison of hemodynamic changes in a chronic nonpulsatile biventricular bypass (BVB) and total artificial heart (TAH). Trans Am Soc Artif Intern Organs. 1980;26:455–460
25. Saito S, Nishinaka T. Chronic nonpulsatile blood flow is compatible with normal end-organ function: implications for LVAD development. J Artif Organs. 2005;8:143–148
26. Tsukiya T, Tatsumi E, Nishinaka T, et al. Design progress of the ultracompact integrated heart lung assist device–part 1: effect of vaned diffusers on gas-transfer performances. Artif Organs. 2003;27:907–913
27. Boschetti F, Cook KE, Perlman CE, Mockros LF. Blood flow pulsatility effects upon oxygen transfer in artificial lungs. ASAIO J. 2003;49:678–686
28. Hormes M, Borchardt R, Mager I, Rode TS, Behr M, Steinseifer U. A validated CFD model to predict O2 and CO2 transfer within hollow fiber membrane oxygenators. Int J Artif Organs. 2011;34:317–325
29. Siggard-Andersen O, Wimberly PD, Fogh-Andersen N, Gøthgen IH. Measured and derived quantities with modern pH and blood gas equipment: Calculation algorithms with 54 equations. Scand J Clin Lab Invest. 1988;48(Supp 189):7–15
Copyright © 2012 by the American Society for Artificial Internal Organs