Peripheral access cardiopulmonary bypass (CPB) is initiated with percutaneous cannulae (CTRL) and venous drainage is often impeded due to smaller vessel and cannula size. A new cannula (Smartcanula ®, SC) was developed which can change shape in situ and, therefore, may improve venous drainage. Its performance was evaluated using a 2-D computational fluid dynamics (CFD) model. The Navier-Stokes equations could be simplified due to the fact that we use a steady state and a 2-dimensional system while the equation of continuity (ρ constant) was also simplified. We compared the results of the SC to the CTRL using CFDRC® (Version 6.6, CFDRC research corporation, Huntsville, USA) at two preloads (300 and 700 Pa). The SC's mass flow rate outperformed the CTRL by 12.1% and 12.2% at a pressures of 300 and 700 Pa, respectively. At 700 Pa, a pressure gradient of 50% was measured for the CTRL and 11% for the SC. The mean velocity at the 700 Pa for the CTRL was 1.0 m.s-1 at exit while the SC showed an exit velocity of 1.3 m.s-1. Shear rates inside the cannulae were similar between the two cannulae. In conclusion, the prototype shows greater mass flow rates compared to the classic cannula; thus, it is more efficient. This is also advocated by a better pressure gradient and higher average velocities. By reducing cannula-tip surface area or increasing hole surface area, greater flow rates are achieved. Perfusion (2007) 22, 257—265.

Central venous catheters (CVC) have been in wide used as vascular access in hemofiltration. The demand of high flow rates for CVC has increased in the era of high-flux dialiysis. The new double lumen Smart Cat prototype is based on a self expanding structure which keeps the vessel open. Hence, SmartCat CVC will provide a lower arterial and venous flow resistance. For the Smart Cat CVC design phase, pressure drop values for different arterial and venous geometries and lumen sizes were simulated using CFD. Rercirculation problems were also simulated using a CFD approach. First prototypes were manufactured based on CFD results and tested in a benchmark. We have measured and compared pressure drop (dP)/blood flow rates (Q) in Smart Cat 11.5F 15cm (n=3) vs. a concentric double lumen Standard CVC 11.5F 15cm (n=3). Arterial (outflow) and venous (inflow) pressure values were measured (dPa, dPv) for different Q. dP and Q were measured at pump speeds from 100–600 ml/min in increments of 100 ml/min. Experiments were carried out using water as liquid phase. CFD simulations during development process allowed to optimize the geometry of the CVC. In vitro results shows the outstanding performance of the new Smar Cat CVC when compared to standard CVC. Smart Cat allows higher blood flow rates for a given pressure drop, which permits shorter blood filtration sessions.Table: experimental results

We report on our efforts using computational fluid dynamic (CFD) tools to compute the O2 transfer capacity of HFGTD. We propose a 3D model to describe O2 transfer in a reduced number of HFs regularly or randomly distributed. During last years, we have developed an intravenous HFGTD as a potential device for the treatment of acute respiratory problems. Comparing with extracorporeal HFGTD where HFs are regularly disposed, we describe our prototype as a randomly distributed bundle of HFs. We measured using in-vitro tests, the gas transfer capacity of our prototypes (VO2=47.2±3, VCO2=69.7±6 ml/min). Tests were run with different blood and gas flow rates (Qb=1–4 l/min, Qg=3–6 l/min). We assess from our data, that VO2 value in our devices is still limited. We presume the channelling effect due to random distribution of HF might decrease their O2 transfer capacity. A computational model is used to study the role of HF distribution on O2 transfer capacity. Fluid equations governing blood motion and advection-diffusion equation for O2 transport are solved using a CFD code (Fluent). Appropriate boundary conditions (BC) for blood motion and O2 transfer are computed. Inlet velocity is set for different Qb, O2 transfer BC through HF wall and surrounding fluid is fixed using experimentally computed mass transfer coefficients (3.1l<6.8*10-5 m/s). The CFD results indicate that channelling through randomly packed bundles can reduce O2 transfer capacity relative to regularly packed hollow fibre bundles. CFD offers a detailed understanding of blood motion and O2 transfer in HFGTD.

When lung function is compromised, alternative devices need to be deployed in order to maintain blood oxygenation. A new device, NovaLung ® , has been designed for acute lung failure. We went about evaluating its gas exchange capability. Three calves (79.5±7.8 kg) were connected to the NovaLung ® System with a priming volume of 240 mL, gas exchange surface area of 1.3 m 2 and exhibiting a biologically coated surface. A standard battery of blood samples were taken before implantation and over a six hour period. Hematocrit remained stable ranging from 27±4% (baseline) to 29±5% (6 hrs). Platelets were preserved ranging from 882±27.4 U/L (baseline) to 734±147 (6 hrs). LDH remained stable at 719±85 U/L (baseline) vs 686±190 U/L (6 hrs) and the pressure drop was maintained below 20 mmHg. Minimal hemolysis was observed. Oxygen transfer peaked at two hours acute extracorporeal lung support (ECLS) with a mean value of 130±50 ml/min. In conclusion, the device is easy to use, provides adequate O 2 and CO 2 transfer for partial lung support in an acute setting. Shows minimal signs of hemolysis and platelets levels are maintained throughout the six hour ECLS period.

INTRODUCTION: Computational fluid dynamics (CFD) techniques are becoming more and more mainstream in the cardiovascular research. It is an important tool to better understand blood flow characteristics, wall shear stresses, and recirculation zones in the arterial wall regions which are usually very difficult information to obtain in vivo. However providing these detailed descriptions on a patient basis is very complex, in particular due to the wide variety of individual vascular morphologies.

Devices for venous cannulation have seen significant progress over time: the original, rigid steel cannulas have evolved toward flexible plastic cannulas with wire support that prevents kinking, very thin walled wire wound cannulas allowing for percutaneous application, and all sorts of combinations. In contrast to all these rectilinear venous cannula designs, which present the same cross-sectional area over their entire intravascular path, the smartcanula concept of "collapsed insertion and expansion in situ" is the logical next step for venous access. Automatically adjusting cross-sectional area up to a pre-determined diameter or the vessel lumen provides optimal flow and ease of use for both, insertion and removal. Smartcanula performance was assessed in a small series of patients (76 +/- 17 kg) undergoing redo procedures. The calculated target pump flow (2.4 L/min/m2) was 4.42 +/- 61 L/ min. Mean pump flow achieved during cardiopulmonary bypass was 4.84 +/- 87 L/min or 110% of the target. Reduced atrial chatter, kink resistance in situ, and improved blood drainage despite smaller access orifice size, are the most striking advantages of this new device. The benefits of smart cannulation are obvious in remote cannulation for limited access cardiac surgery, but there are many other cannula applications where space is an issue, and that is where smart cannulation is most effective.

Methods: Thirty-nine commercially available PTFE tubular prostheses provided by different companies (Atrium, Gore-Tex and Impra) in three different diameters (4, 5 and 6mm) have been mounted on a system allowing filling with H2O at the wanted pressure and measurement of the internal diameter with intra-vascular ultrasound (Clear View Ultra® System, Boston Scientific, Sunnyvale, CA). For each prosthesis (3cm length) 4 diameters were measured respectively at 0, 30, 60 and 90° of the circumference, and the measurements repeated at three different intervals corresponding to the middle and each extremity. All 468 measurements were recorded with an internal pressure=100mmHg. Correspondent flows for a pressure drop from 100 to 15mmHg were calculated.

In this work we report on recent efforts in our laboratory to couple experimental and computational approaches to analyze the gas transfer performance of an IHFGTD. We describe a 3D computational approach to characterize blood motion and O25 transfer in a simplified IHFCTD. Modelling each hollow fiber (HF) of the device by a 3D numerical model is an enormous task. We propose a simplified computational model to analyze blood flow and O2 transfer between HF and blood. This is based on a reduced number of crimped HF (N = 20) regularly distributed. Experimental approach using a saline solution is also carried out to calculate gas transfer performance (VO,) of our prototypes for several inlet conditions. Then, we compute the mass transfer coefficient (VO2) with blood taking the equation K1=αReβSc1/3 and using the procedure proposed by Mockros. These results are derived using 3 IHFCTD protatypes placed in an in-vitro set up. Their gas exchange area is 0.5 m2. Navier-Stokes equations for blood flow and advection-diffusion for O2 transfer are solved using a CFD code (Fluent, Lebanon, NH). Appropriate boundary conditions (BC) are set for blood motion and O2 transfer in blood. The fluid is set as steady, laminar, incompressible and it is assumed to be Newtonian. Inlet condition is set for different blood flow rates (1 ≥Qb≥3 1/min). Robin BC for O2 transfer is set using experimentally deduced mass transfer coefficients (3.1≥K1≥6.8*10−5 m/s). CFD techniques offers a detailed understanding of blood mechanics and O2 transfer for medical devices.

Background. The purpose of this paper was to evaluate myocardial and pulmonary effects of aqueous oxygen (AO) delivered directly into the pulmonary circulation in acute hypoxia.Methods. Six calves (2 months old, 68.0 +/- 2.2 kg) after general anesthesia, mechanical ventilation, and median sternotomy underwent total right heart bypass using fixed flow with continuous pressure and blood gas measurements in carotid and femoral arteries, left atrium, the coronary sinus and PA. Measurements of systemic and PA pressures and O-2 saturations; myocardial O-2 atrioventricular (AV) differences; and 02 extraction were made. After base line measurements, hypoxic ventilation reducing the mean arterial PO2 from 277 +/- 102 nun Hg to 47 +/- 4 mm Hg (p < 0.0005) was maintained for 30 minutes. Without changes in the hypoxic ventilation (mean arterial PO2 = 49 +/- 11 mm Hg) 3 ml/min of hyperbaric aqueous oxygen (AO = oxygen diluted in saline solution) was administered into the PA for 30 minutes. Pulmonary blood flow was maintained during the entire experiment (3.7 +/- 0.3 L/min).Results. Hypoxic ventilation significantly raised (p < 0.05) the systolic (30 +/- 7 vs 21 +/- 4 mm Hg), diastolic (20 +/- 6 vs 12 +/- 3 rum Hg), and mean (23 +/- 7 vs 15 +/- 3 mm Hg) PA pressure; PA/systemic pressure ratio for systolic (0.37 +/- 0.08 vs 0.25 +/- 0.06) and mean (0.56 +/- 0.19 vs 0.29 +/- 0.11) pressures; and pulmonary vascular resistance (PVR) (5.63 +/- 1.06 vs 3.53 +/- 0.73 U). Aqueous oxygen (AO) infusion significantly reduced (p < 0.05) the values obtained with hypoxic ventilation; systolic (23 +/- 5 vs 30 +/- 7 mm Hg), diastolic (11 +/- 4 vs 20 +/- 6 mm Hg), and mean (14 +/- 3 vs 23 +/- 7 mm Hg) PA pressure; PA/systemic pressure ratio for systolic (0.25 +/- 0.05 vs 0.37 +/- 0.08) and mean pressures (0.29 +/- 0.12 vs 0.56 +/- 0.19); and PVR (3.41 +/- 1.01 vs 5.63 +/- 1.06 U). AO infusion in the pulmonary circulation did not influence the myocardial O-2 atrioventricular (AV) difference or the O-2 extraction.Conclusions. Infusion of hyperbaric AO solution into the PA can completely reverse the negative effects of acute hypoxia on the pulmonary circulation without affecting the myocardial metabolism. (C) 2004 by The Society of Thoracic Surgeons.

Objective: To evaluate the effects of hyperbaric oxygen solution on hypoxic pulmonary hypertension.Methods: Eleven calves, 2-month-old, 71 +/- 6 kg, underwent general anaesthesia, mechanical ventilation and median sternotomy. Catheters for continuous pressure and blood gas measurements were inserted in carotid and femoral arteries, left atrium, right atrium and pulmonary artery (PA), and a flow-probe placed around the PA. After baseline measurements 30 min hypoxic ventilation reduced the mean arterial PO2 from 285 +/- 115 to 46 +/- 11 mmHg (P < 0.0001). At this point, without changes in hypoxic ventilation (mean arterial PO2 maintained at 50 5 mmHg), 3 ml/min of hyperbaric aqueous oxygen (AO, oxygen diluted in saline solution) was infused directly into the PA for 30 min, with continuous reading of the monitored parameters.Results: Hypoxic ventilation raised significantly (P < 0.005) the values of systolic (36 +/- 7 vs 22 +/- 6 mmHg), diastolic (16 +/- 3 vs 9 +/- 4 mmHg) and mean (24 +/- 4 vs 14 +/- 4 mmHg) PA pressure, PA/systemic pressure ratio for systolic (0.47 +/- 0.09 vs 0.24 +/- 0.06) and mean (0.49 +/- 0.13 vs 0.23 +/- 0.08) pressures and Pulmonary Vascular Resistance (PVR) (6.89 +/- 0.87 vs 2.67 +/- 0.38 U), while the Pulmonary Blood Flow (PBF) decreased (2.7 +/- 0.4 vs 3.7 +/- 0.4 l/min). AO infusion reduced significantly (P < 0.005) the values obtained with hypoxic ventilation with systolic (26 +/- 6 vs 36 7 mmHg), diastolic (11 +/- 4 vs 16 +/- 3 mmHg) and mean (16 +/- 4 vs 24 +/- 4 mmHg) PA pressure, PA/systemic pressure ratio for systolic (0.27 +/- 0.07 vs 0.47 +/- 0.09) and mean (0.27 +/- 0.08 vs 0.49 +/- 0.13) pressures and PVR (3.42 +/- 0.31 vs 6.89 +/- 0.87 U), while the PBF increased (3.6 +/- 0.4 vs 2.7 +/- 0.4 l/min).Conclusions: Acute infusion of hyperbaric AO solution into the PA completely reverses the negative effects of acute hypoxia on pulmonary circulation. (C) 2004 Elsevier B.V. All rights reserved.

In this study we report on recent efforts in our laboratory to couple experimental and computational approaches to analyze the O2 transfer (VO2) of a IHFO. We describe two 3D numerical models (3D-NM) to characterize the fluid dynamics and VO2 in a IHFO. Experimental data permit to calculate VO2 and main parameters used for numerical approach. Mass transfer coefficient (K) in terms of Reynolds (Re) and Schmid (Sc) numbers are derived from Mockros proposed equation: (1) K∼α*Re β*Sc1/3 Relation between pressure drop (ΔP) and flow (Um) is derived using the equation: (2) ΔP= λ*Um+ θ*U2m where δ, ± factors are derived from experimental data. Empirical results were derived using 3 prototypes of crimped IHFO placed in a chamber. Devices membrane area was 0.5m2. Modeling each individual fiber of the IFHO by a 3D-NM remains an intractable task. Two simplified models are proposed: “Fiber-Fluid” (FF) and “Porous-media” (PM)·FF: A reduced number of fibers is considered as fluid domain. It permit to analyze in detail fluid-fiber iteration and mass transfer. • PM: It permit to get a global sight of gas transfer. The IHFO is replaced by a PM with a given permeability derived from equation (2) and fibers are replaced by O2 source terms. Dirichlet or Robin boundary conditions are set for mass transfer between fluid and fibers. Flow is treated as steady, laminar, imcompressible and fluid is assumed to be Newtonian. Continuity and Navier-Stokes governing the flow and Convection-Diffusion for mass transfer equations were solved using a FVM CFD code (Fluent, Lebanon, NH).

A high resolution echo-tracking system permits the calculation of cross-sectional compliance considering vessel diameter variations alone, and assumes that longitudinal movement of the vessel wall due to pulse pressure is negligible. However, using piezoelectric crystals sutured on the adventitia of the vessel wall we demonstrated that arterial length changes up to 5% (mean 2.7%) as a function of pulse pressure. Therefore, cross-sectional compliance seems to provide a limited approximation of the real phenomenon because it neglects axial vessel movement. Axial vessel movement is taken into account when the vessel compliance is calculated according to the principle of continuity of the mass: [equation: see text]. To verify this hypothesis we measured the blood flow gradient through 10 cm long segments of 10 pig carotid arteries (Qin - Qout) and divided it for the derivative of blood pressure over a given time (deltaP/deltat). For the same vessels, we calculated the cross-sectional compliance (CC) using the echo-tracking system (NIUS 02). We found a CC of (5.91 +/- 0.4) x 10(-7) micro m(2)/mm Hg and a segmental carotid compliance or dynamic compliance (C(d)) of (6.21 +/- 0.2) x 10(-8) micro m(3)/mm Hg. The impact of axial strain in calculations of compliance results in a dynamic compliance, which is one order of magnitude smaller than traditionally calculated arterial compliance.

To demonstrate axial artery motion during the cardiac cycle, the common carotid arteries (CCA) of 10 pigs were exposed and equipped with piezoelectric crystals sutured onto the artery as axial position detectors. An echo-tracking system was used to simultaneously measure the CCA diameter. For each animal, data for pressure, length, and diameter were collected at a frequency of 457 Hz. At a mean pulse pressure of 33 +/- 8 mmHg, the mean systolodiastolic length difference was 0.3 +/- 0.01 mm for a mean arterial segment of 11.35 +/- 1.25 mm. Systolic and diastolic diameters were 4.1 +/- 0.3 and 3.9 +/- 0.2 mm, respectively. The examined CCA segment displayed a mean axial systolic shortening of 2.7%. This study clearly demonstrates, for the first time, that the length of a segment of the CCA changes during the cardiac cycle and that this movement is inversely correlated with pulse pressure. It is also apparent that the segmental axial strain is significantly smaller than the diameter variation during the cardiac cycle and that the impact of the axial strain for compliance computation should be further evaluated.

The efficacy of a new polymethylpentene (PMP) hollow fiber membrane (inner diameter of 200μm and wall thickness of 90μm) was tested with an intravenous oxygenator and carbon dioxide removal prototype. The prototypes consisted of a polyurethane potted bundle of crimped hollow fibers with a membrane surface area of 0.38m2. The prototypes (two groups of n=3) gas transfer performance was tested using a buffered cristalloyd solution (pH=7.4) in a completely automated in vitro system. Each one was placed in a plexyglass chamber (inner diameter of 20mm) as host vessel. Oxygen and carbon dioxide transfer values (VO2, VCO2) were calculated and displayed continuosly at different liquid flow rates through the chamber (Qb=1; 2; 3; 5l/min) and gas flow rates (Qg=1; 3; 51/min). Permissive hypercapnia was modelized increasing venous CO2 partial pressure (pCO2=45; 60; 90mmHg). Liquid pressure drop in the chamber was also measured (ΔP mmHg). Gas transfer values were compared at the same conditions, with three other prototypes (with identical membrane surface area) made with polypropylene (PP) hollow fibers (inner diameter of 240μm and wall thickness of 30μm). Maximal and minimal gas transfer mean values are resumed in the following table:TableMaximal mean ΔP (for Qb= 5 l/min) for PP prototypes is 17.3±2.5mmHg and ΔP=28.71±6.1mmHg for those made with PMP. These results demonstrate a better performance of PMP fibers for O2 transfer. Besides, experiments showed an adequate capacity for CO2 transfer. Their smaller inner diameter and the higher hollow fiber densitiy in the chamber (derived from their bigger outer diameter) are effective ways to increase O2 transfer.

Commonly, oxygenators are evaluated using in-vivo (bovine or porcine) models. This is a cumbersome an expensive procedure. Our objective was to develop a completely automated saline solution or blood based in-vitro evaluation system to measure the performance of gas transfer devices. We used this novel system to evaluate gas transfer performance of a commercially available cardio-pulmonary bypass oxygenator. The Baxter Spiral GoldTM oxygenators were tested using bovine blood according to AAMI standards. The circuit consisted of two bubble oxygenators arranged in parallel, fed by two mass flow controllers with N2 and CO2 to deoxygenate and replenish the solution with CO2 to achieve desired inlet conditions. Fine control of the test oxygenator venous inlet conditions were achieved electronically using in-line blood gas monitoring and electronic N2 and CO2 mass flow controllers. Oxygenator pressure drop, VO2 and VCO2 gas transfer rates were calculated and displayed using a computerized data acquisition system. The gas transfer data we obtained for this oxygenator correlated well with the manufacturer's published values. For a blood flow rate of Qb=3 L/min, Qg:Qb =1 and FiO2 =1, the gas transfer rates were VCO2=162.8 ±10 mL/min and VO2=227.46 ±5 mL/min. These results demonstrate the viability of our in-vitro system for evaluating oxygenator efficacy without the use of an animal or surgical procedure.

The effects of diaspirin crosslinked hemoglobin (DCLHb, Baxter Health Care Corp., Round Lake, IL) on oxygen exchange in the setting of cardiopulmonary bypass (CPB) are unknown. Six calves (71.2 +/- 1.3 kg) were connected to CPB by jugular venous and carotid arterial cannulation for 5 hours. Each 1 hour period included 45 min of partial CPB (mean flow rate of 50 ml/kg per min) followed by 15 min without CPB, at the end of which 500 ml of blood were substituted for with either 500 ml of hydroxyethyl starch (Haes; n = 3) or 500 ml of DCLHb (n = 3). A total of 2 liters of blood was, thus, exchanged (28 ml/kg of blood substitute). Values are expressed as mean +/- 1 SD. Analysis of variance for repeated measurements was used. The cardiac output (CO) values at 1 h, 3 h, and 5 h were in the Haes group: 5.7 +/- 2, 6.7 +/- 2.5, and 7.7 +/- 2.5L/min, and in the DCLHb group: 5.7 +/- 0.6, 4 +/- 1, and 4.7 +/- 1.2 L/min, respectively. The arteriovenous oxygen content difference (Ca-Cvo2) values at 1 h, 3 h, and 5 h were in the Haes group: 4.6 +/- 1, 3.3 +/- 1.5, and 3.5 +/- 1.5 ml/dl, and in the DCLHb group: 4.9 +/- 0.6, 7.4 +/- 0.7, and 6.6 +/- 0.6 ml/dl, respectively. The oxygen consumption (Vo2) values at 1 h, 3 h, and 5 h were in the Haes group: 244 +/- 29, 198 +/- 58, and 249 +/- 42 ml/min, and in the DCLHb group: 273 +/- 28, 296 +/- 75, and 306 +/- 65 ml/min, respectively. CO and Ca-Cvo2 showed a significant difference (p < 0.01), whereas Vo2 did not (p = 0.52). In the DCLHb group of this CPB animal model, the cardiac output is lower and the arteriovenous oxygen content difference higher than in the Haes group, allowing for preserved oxygen consumption.

Background: In our hands, in vivo segmental vessel length changes up to 5% because of blood pressure: increasing in arterial pressure is associated to decrease in segmental vessel length. Methods and Material: Using two piezoelectric crystals sutured on vessel wall and a high fidelity pressure probe, we recorded artery length variations as function of blood pressure, before and after an end-to-end anastomosis on four pigs carotid arteries. Results: Mean arterial pressure before anastomosis = 73 mmHg (+/- 12); mean arterial pressure after anastomosis = 91 mmHg (+/- 14); mean crystals displacement before anastomosis during systole = -0.21mm; mean crystals displacement after anastomosis during systole = +0.24 mm; mean distance between crystals before anastomosis = 12.3 mm (+/- 0.8) and after anastomosis = 11.2 mm (+/- 0.5). Conclusions: In the acute phase following an end-to-end anastomosis, an increase in blood pressure causes increasing in vessel length, with an exponential correlation. The anastomosis is constantly subjected to a longitudinal traction whose magnitude depends on blood pressure.

The information gathered with intravascular ultrasound (IVUS) are of great value in endovascular techniques. The aim of this study was to evaluate the reliability of IVUS when measuring vessel dimensions by comparison with an established reference method. The left carotid artery was exposed in 4 pigs (45-55 kg) and two piezoelectric crystals were sutured on the adventitia in the same cross-sectional plane. The distance between them was measured either by IVUS and by sonomicrometers. The mean distance between the two crystals calculated by the sonomicrometer was 4.7 +/- 0.4 mm (mean systolic distance was 4.9 +/- 0.2 mm, mean diastolic distance was 4.6 +/- 0.1 mm). The mean distance between the two targets calculated by IVUS was 4.5 +/- 0.2 mm (mean systolic distance was 4.6 +/- 0.2 mm and mean diastolic 4.4 +/- 0.2 mm). Regression analysis of the two series of data shows a R-2 = 0.9984. IVUS measurements are an average 5% smaller than sonomicrometer measurements (3.6% up to 8.3%) and the difference is statistically significant (p<0.05). The underestimation of IVUS measurements will affect the accuracy, and probably the long-term outcome, of endovascular procedures.

Several manufacturers offer heparin-coated cardiopulmonary bypass (CPB) circuits for improved biocompatibility. However, the effect that heparin treatment has on the efficacy of oxygenator gas transfer is seldomly addressed. The objective of this study was to elucidate the effect of heparin surface treatment on the gas transfer efficacy of polypropylene hollow fiber membrane oxygenators. We compared the oxygen and carbon dioxide transfer efficacy of two different heparin-coated CPB oxygenators with uncoated oxygenators of the same design. The oxygenators we evaluated were the 3M Sarns TurboTM, and the Baxter Duraflo Spiral GoldTM. The study was performed using a novel in-vitro saline based evaluation system with electronically controlled blood and gas flows, in-line blood gas monitoring, and computerized data acquisition. A hollow fiber membrane oxygenator was connected in series with the test oxygenator and supplied with N2 and CO2 for deoxygenation and CO2 replenishment to achieve AAMI inlet conditions. The evaluations were performed with flows of 2, 3, and 4 L/min (Q gas = Q fluid), and FiO2 = 1. Our results indicate that heparin coating has negligible effect on O2 transfer, and moderate effect on CO2 transfer to saline solution. The 3MTM heparin-coated oxygenators displayed only slightly reduced O2 (74.4±0.63 vs. 75.8±0.54 ml/min) and more significantly reduced CO2 (71.5±9.38 vs. 106±38.3 ml/min) transfer rates at 4 L/min flow rate. These trends were consistent across the range of flow rates. The BaxterTM heparin-coated oxygenators displayed slightly increased O2 (72.0±0.48 vs. 71.2±0.57 ml/min) and moderately reduced CO2 (116.4±12.6 vs. 127.4±3.2 ml/min) transfer rates at 4 L/min flow rate. These trends were also consistent across the range of flows.