Experimental use of a compact centrifugal pump and membrane oxygenator as a cardiopulmonary support system

Compactness and high performance are the most important requirements for a cardiopulmonary support system. The Nikkiso (HPM-15) centrifugal pump is the smallest (priming volume; 25 ml, impeller diameter; 50 mm) in clinically available centrifugal pumps. The Kuraray Menox (AL-2000) membrane oxygenator, made of double-layer polyolefin hollow fiber, has a minimum priming volume (80 ml) and a low pressure loss (65 mm Hg at 2.0 L/min of blood flow) compared with other oxygenators. The aim of this study was to evaluate the performance of the most compact cardiopulmonary support system (total priming volume: 125 ml) in animal experiments. The cardiopulmonary bypass was constructed in a canine model with the Nikkiso pump and Menox oxygenator in comparison with a conventional cardiopulmonary support system. The partial cardiopulmonary bypass was performed for 4 h to evaluate the gas exchange ability, blood trauma, serum leakage, hemodynamics, and blood coagulative parameters. The postoperative plasma free hemoglobin level of the compact cardiopulmonary system was 29.5 +/- 10.21 mg/dl (mean +/- SD), which was lower than that of the conventional cardiopulmonary system, 48.75 +/- 27.39 mg/dl (mean +/- SD). This compact cardiopulmonary system provided the advantage in terms of reduction of the priming volume and less blood damage. These results suggested the possibility of miniaturization for the cardiopulmonary bypass support system in open-heart surgery in the near future.     

Initial In Vitro Evaluation of a Pediatric Vortex-Mixing Membrane Lung

Abstract: A new design for a pediatric membrane lung is described in this paper. The lung consists of eight blood compartments, each having six U-shaped blood channels, with microporous PTFE membranes supported on rigid plates in such a way that the membranes form furrowed blood channels. Two rolling diaphragm pumps are attached to the open ends of the U-shaped blood channels; these pumps are operated in antiphase. Mean flow is provided by a roller pump placed at the inlet end of the membrane lung. Pulsatile blood flow within the blood channels produces successive vortex formation and ejection, leading to good blood mixing and high efficiency in gas transport. The design of the rolling diaphragm piston pumps ensures that the blood prime volume is low (280 ml), and the grouping of the pumps at one end of the oxygenator allows the driving mechanism to be simple and compact. The relatively wide blood channels (minimum width 0.5 mm) and vortex mixing make priming the membrane lung particularly easy. The membrane area is 0.39 m². Preliminary performance testing of the pediatric membrane lung was undertaken by pumping blood around a circuit containing a roller pump, the membrane lung, and a bubble oxygenator (to adjust the blood gases at the inlet to the membrane lung). In five such experiments it was shown that the membrane lung transferred 80 ml O₂/min and 120 ml CO₂/min at a blood flow rate of 1.5 L/min.     

Study of Regional Cerebral Oxygen Saturation During Percutaneous Cardiopulmonary Support

Abstract: The purpose of this study was to evaluate the change of regional cerebral oxygen saturation (rSo₂) during percutaneous cardiopulmonary support (PCPS) in patients with cardiogenic shock. Fifteen patients with cardiogenic shock were evaluated during PCPS by continuous monitoring of rSo₂, systemic venous oxygen saturation (Svo₂), and hemodynamics. The brain damage of these patients was also evaluated during and after PCPS. There were 10 males and 5 females. Their ages ranged from 57 to 79 years old (average: 60.0 ± 14). Two patients were unconscious before PCPS, and 11 received intraaortic balloon pumping (IABP) before PCPS. The change of rSo₂was significantly correlated with the change of Svo,. The average of rSo₂was 64 ± 3% at the stable hernodynamic condition. The rSo₂with pulsatile PCPS was higher than that with nonpulsatile PCPS. There was no correlation between brain damage and rSo₂. The patients with low rSo₂(40%) that resulted in poor LV function could not be weaned from PCPS. In conclusion, the continuous monitoring of rSo₂during PCPS could be a useful tool.     

Clinical experience of a percutaneous ventricular support for profound heart failure following acute myocardial infarction--a case report]

Percutaneous biventricular support with centrifugal pump was applied to a 59-year-old man with profound heart failure following acute myocardial infarction. Direct coronary angioplasty was performed under percutaneous cardiopulmonary support (PCPS). After angioplasty, he could not be weaned from PCPS under IABP. We used percutaneous left ventricular support (PLVS) with transseptal left atrial cannulation. PLVS could not maintain effective assist flow and he developed a cardiogenic shock again. PCPS was added to PLVS, and percutaneous biventricular support was started. Assist flow ranged 1.7-3.0 L/min in PLVS, and 1.7-2.0 L/min in PCPS, and total assist flow was 3.4 L/min. Pulsatile pressure was obtained by adding IABP. The patient was successfully weaned from PCPS in 26 hours and PLVS in 118 hours. Patient is alive and well one year after PLVS. Combination of PLVS and PCPS is an effective ventricular support system, being less invasive than conventional method with thoracotomy, and may be useful for profound heart failure.     

Percutaneous Cardiopulmonary Support as the Second Generation of Venoarterial Bypass: Current Status and Future Direction

Abstract: We have developed a new percutaneous cardiopulmonary support (PCPS) system in cooperation with Terumo Corporation, called the Emergency Bypass System (EBS). This preassembled system is unique in terms of its small priming volume (470 cc), fully automatic priming function, originally developed membrane oxygenator, and Capiox straight path centrifugal pump. The priming process takes only 5 min. We have used this system in 4 patients (all were male; mean age 56 years, range 35–73 years). The duration of assist ranged from 15 min to 210 h. Maximum bypass flow ranged from 2.2 to 3.6 L/min, systemic circulation was very well maintained, and urinary output was acceptable. Although the centrifugal pump was in good working condition, we had to change the oxygenator because of serum leakage 36 to 48 h after initiation of the assist. One patient was weaned from the PCPS, but died of brain death. The other patients could not be weaned from the PCPS. The cause of death in these patients was irreversible myocardial damage. Disseminated intravascular coagulopathy triggered by surgical trauma related to various procedures developed in 2 patients during assist and was the cause of cessation of the assist. Problems to be ratified and future directions in the EBS and PCPS were earlier introduction through the establishment of clear indications and endpoint, heparin coating, improvement in durability of the oxygenator, and cost.     

Thoughts and Progress

In recent years, extracorporeal membrane oxygenation (ECMO) has been used for treatment of neonates with respiratory failure. A prototype of a compact ECMO system for neonates was developed. A single-lumen catheter, inserted into the right atrium via a jugular vein, was used for withdrawal and infusion of blood through the catheter. An extracapillary flow hollow-fiber membrane lung made of microporous polypropylene has a total surface area of 0.6 m2. To prevent the increase of plasma free hemoglobin, the ratio of withdrawal/infusion is controlled by a microcomputer. The system is compact in size with a low priming volume (less than 90 ml), which allows for ECMO with no additional blood transfusions. Its potential application as a respiratory support system is evaluated in animal experiments. The total intermittent veno-veno bypass flow was 15-30 ml/min/kg. The O2 transfer rate was 20 ml/min and the CO2 transfer rate was 33 ml/min at a blood flow rate of 300 ml/min. The O2 and CO2 exchange with the ECMO system was efficient enough to eliminate the respiratory failure induced by mechanical ventilation. The increase in plasma free hemoglobin was only 4 mg/dl after 6 h of ECMO. The system was considered applicable to respiratory aid for neonates.     

Flow rate and pressure head estimation in a centrifugal blood pump

Considering a miniaturization of percutaneous cardiopulmonary support (PCPS) system, we have developed a pump system that can estimate flow rate and pressure head in the circuit. A commercially available centrifugal blood pump (C1E3, Kyocera Co., Ltd., Kyoto, Japan) with an exclusive motor driver was used in the experiment. In this system, the motor revolutions per minute and torque were used to derive an estimation equation. Taking into consideration fluid viscosity, the accuracy of the estimation was increased. Bovine blood was used to evaluate the system in vitro. The average maximum difference between the estimated flow rate and the measured flow rate was 0.56 L/min (for 98% accuracy). The average difference between the estimated pressure head and the measured pressure head was 30.7 mm Hg (for 94.2% accuracy). Because of the stability of the estimations, we believe this system has the possibility of use for real-time monitoring and miniaturization of PCPS system.     

Development of an Ultracompact Integrated Heart-Lung Assist Device

A novel integrated heart-lung assist device has been developed as a simple to use portable cardiopulmonary support system. The device comprises a centrifugal pump and an artificial lung, which is located around the pump, in an all in one system. The special membrane employed precludes plasma breakthrough in protracted use and enables preprimed setup. Test lungs consisting of the same membrane preserved gas exchange function well after 3 months of preprimed storage. The entire blood contacting surface is treated with covalent heparin bonding to impart good antithrombogenicity. Heparin bonded test lungs could be continuously perfused without systemic anticoagulation as long as 36 days in a venoarterial bypass chronic animal study using goats. The prototype device (diameter, 126 mm; height, 59 mm; membrane area, 0.85 m2; priming volume, 180 ml) demonstrated 9 L/min pump output at a 400 mm Hg pressure head and 180 ml/min oxygen and 110 ml/min carbon dioxide transfer rates at 5 L/min blood flow. We conclude that this device has potential to be the next generation cardiopulmonary support system.     

Effects of Ultrathin Silicone Coating of Porous Membrane on Gas Transfer and Hemolytic Performance

Abstract: To assess the effect of an ultrathin (0.2 μm) silicone-coated microporous membrane oxygenator on gas transfer and hemolytic performance, a silicone-coated capillary membrane oxygenator (Mera HP Excelung-prime, HPO-20H-C, Senko Medical Instrument Mfg. Co., Ltd. Tokyo, Japan) was compared with a noncoated polypropylene microporous membrane oxygenator of the same model and manufacturer using an in vitro test circuit. The 2 oxygenators showed little difference in the oxygen (O₂) transfer rate over a wide range of blood flow rates (1 L/min to 8 L/min). The carbon dioxide (CO₂) transfer rate was almost the same in both devices at low blood flow rates. but the silicone-coated oxygenator showed a decrease of more than 20% in the CO₂ transfer rate at higher blood flow rates. This loss in performance could be partly attenuated by increasing the gas/blood flow ratio from 0.5 or 1.0 to 2.0. In the hemolysis study, the silicone-coated membrane oxygenator showed a smaller increase in plasma free hemoglobin than the noncoated oxygenator. The pressure drop across both oxygenators was the same. These results suggest that the ultrathin silicone-coated porous membrane oxygenator may be a useful tool for long-term extracorporeal lung support while maintaining a sufficient gas transfer rate and causing less blood component damage.     

Clinical Experience of Percutaneous Cardiopulmonary Support

Abstract: Recently, percutaneous cardiopulmonary support (PCPS) combined with femoro-femoral bypass without reservoir has become valued because of its quick and easy application. We developed a fully preconnected compact integrated cardiopulmonary bypass (CPB) unit (priming volume of 250 ml) with a blind pore membrane oxygenator (Kuraray Menox) for PCPS. From 1990 to 1995, PCPS was performed in 49 patients of whom 26 were weaned from support. In most cases, we applied this CICU in patients with no active bleeding (22 patients); in patients with active bleeding (n = 13), we used Medtron-ic's heparin-bonded close chest support pack (CCSP). Of these, PCPS was performed uneventfully for 2 h (median) in 8 elective cases; all of these patients were weaned or were switched to a left ventricular assist system (LVAS). In 8 urgent cases, such as those with low cardiac output syndrome, PCPS was performed for 4 days (median), 1 was weaned, and 2 CICU were cases switched to other procedures. In 32 cases of shock, 5 CICU patients were weaned, and 3 of them survived. Eight patients including 5 CICU patients and 1 CCSP patient were switched to operation or LVAS, and 2 CICU patients remain alive. From these data, PCPS has been shown to support the patient's circulation in the acute phase and earn time to switch to operation or LVAS; the quick and easy set-up of the CICU can improve the clinical results. The use of the Medtronic device broadened the indication for PCPS. The CCSP enlarged the indication of PCPS but could not improve the results. To improve the results, a heparin-bonded surface is desired.     

Intraarterial Perfusion of the Hindlimb with Pyridoxalated Hemoglobin Polyoxyethylene Conjugate Solution in Anesthetized Dogs

Abstract: To evaluate the potential clinical usefulness of a modified hemoglobin, pyridoxalated hemoglobin polyoxyethylene conjugate (PHP), the hindlimb vascular bed was perfused with PHP solution while monitoring tissue oxygen tension (Pto₂) in anesthetized dogs. The hindlimb region was perfused through the external iliac artery with a roller pump at a varying perfusion rate. Pto₂ was measured using a Po₂-monitoring probe inserted into the gra-cial muscle. After surgical preparation for perfusion, the iliac arterial flow rate was 19.9 ± 5.6 ml/min, and baseline Pto₂ was 38.4 ±1.3 mm Hg. Perfusion with autologous arterial blood with the pump increased Pto₂ and perfusion pressure (PP) in a perfusion rate-dependent manner. Perfusion with PHP solution at 20 ml/min decreased Pto₂ from the initial baseline level, but an increase in the flow rate to 40–55 ml/min restored or induced an elevation of Pto₂. Results demonstrated that PHP solution can deliver oxygen to local tissue and maintain tissue oxygen tension at the same level as autologous arterial blood at a high enough flow rate.     

Laboratory evaluation of a new membrane oxygenator with a built-in hemoconcentrator

In order to facilitate the handling of cardiopulmonary bypass (CPB) and simplify the circuit, we have developed a new membrane oxygenator with a hemofiltration function. The hollow fiber units for gas exchange and hemofiltration were combined in concentric circles in a cylindrical housing. The total priming volume was 190 ml. Because we used a silicon-coated hollow fiber membrane for gas exchange, this oxygenator was completely resistant to serum leakage. The gas exchange and hemofiltration sections both have a blood-outside flow configuration. All blood flows in a radial direction from around the central core to the surrounding hollow fiber units, first to the hemofiltration portion and then to the gas exchange section. Filtered fluid was easily collected through a stopcock mechanism. The oxygen transfer rate was 312 ml/min at a blood flow rate of 6 L/min, and the ultrafiltration rate was 3.5 L/hour at a blood flow rate of 4 L/min with 25% hematocrit and 200 mmHg transmembrane pressure in an in vitro study. The pressure drop was 62 mmHg at a blood flow rate of 4 L/min. We found no adverse effects in an in vivo study using a mongrel dog. In conclusion, this durable combined device could achieve excellent and simplified hemoconcentration by having all the blood in the unit flow through the hemofiltration portion, and may be useful not only in CPB during open heart surgery, but also in extracorporeal membrane oxygenation.     

Single-Pass Continuous Flow Peritoneal Dialysis Using Two Catheters

With a renewed interest in continuous flow peritoneal dialysis (CFPD), our standard practice of implanting a second catheter in those patients facing access failure provided us the opportunity to perform acute studies on CFPD in these patients, since it temporarily provided us with two catheters. Four patients were studied, with a total of five studies performed. A standard protocol was followed utilizing 1.5% dextrose solution, a 2 L fill, an inflow rate of 200 ml/min with a proportionate outflow for a 4-hour session. A full drain was performed at the end of the study. Our results provided us with a mean effective peritoneal clearance for urea ( K _peU) and creatinine ( K _peCr) of 40 ml/min and 28 ml/min, respectively, and a mean ultrafiltration rate ( Q _f) of 13.4 ml/min. Our average mass transfer coefficient (MTC) for urea was 40 ml/min, consistent with kinetic modeling and historical data. The K _pe, MTC, and Q _f achieved are significantly higher than other investigators, which could possibly be explained by those obtained by two separate catheters resulting in adequate mixing of the dialysate. These clinical results provide a solid foundation for the future development of this PD modality.     

Long-term application of percutaneous cardiopulmonary support system without systemic heparinization

Though the amount of systemic heparin sodium administration was reduced after the introduction of heparin sodium-coating material to percutaneous cardiopulmonary support system (PCPS), bleeding due to heparin sodium is still the one of the major complications. In 2 patients of postcardiotomy cardiogenic shock, we neutralized heparin sodium by protamine sulfate administration immediately after the institution of PCPS and did not perform systemic heparinization until hemostasis was secured. The time on PCPS without heparin sodium was 37 hours in 1 patient and 91 hours in another patient. While systemic heparin sodium was not administered, activated clotting time (ACT) ranged from 109 to 148 sec and the bypass flow rate was maintained in more than 2.5 l/min. The exchange of the devices was unnecessary during the assistance and the patients were successfully weaned from PCPS without major complications. We conclude that systemic anticoagulation can be avoided in the case of life-threatening hemorrhage.     

The efficacy of low prime volume completely closed cardiopulmonary bypass in coronary artery revascularization.

PURPOSE: This study was conducted to evaluate and demonstrate the efficacy of low prime volume completely closed cardiopulmonary bypass (LPVP) in arrested coronary artery bypass grafting (CABG). We improved the percutaneous cardiopulmonary support (PCPS) circuit to reduce the deleterious effects of cardiopulmonary bypass (CPB). METHODS: Between April 1999 and May 2003, among 228 isolated CABG procedures, 47 procedures using LPVP (group L) and 86 procedures using standard prime volume open CPB (group S) were compared. The LPVP priming volume was 590 mL; the circuit was completely closed with a soft reservoir. Cardiac arrest was obtained by warm blood cardioplegia. RESULTS: The following average values were obtained: packed red blood cell transfusions, 0.88 +/- 1.4 U (group L) vs. 2.1 +/- 2.5 U (group S); intraoperative lowest hematocrit value, 28.7 +/- 4.6% (group L) vs. 22.4 +/- 3.3% (group S); blood loss over first 24 hours, 439 +/- 242 mL (group L) vs. 599 +/- 409 mL (group S); ventilation time, 5.1 +/- 3.1 hours (group L) vs. 10.4 +/- 14.9 hours (group S). CONCLUSION: Compared to standard prime volume open CPB, LPVP resulted in fewer deleterious operative effects. Less blood loss, fewer blood transfusions, and earlier patient recovery was noted with LPVP. Thus, LPVP is a very efficient form of CPB.     

Mechanical Blood Traumatization by Tubing and Throttles in In Vitro Pump Tests: Experimental Results and Implications for Hemolysis Theory

Abstract: Blood has become essential as a test fluid to evaluate hemolysis and biocompatibility of blood pumps in vitro. The blood is usually pumped from a blood bag into a circuit against elevated pressure. A throttle or a length of tubing is used to produce the pressure head. Blood damage caused by the shear stress in these pressure-reducing devices should be minimal. It is not known whether the high but short-lasting shear stress in a throttle is more or less damaging to the blood than the low but long-lasting stress in tubing. In this study, throttles (width 11 mm, minimal height 0.9 mm, length 30 mm; shear stress = 136 N/m² lasting for 3.23 ms); and tubing (inner diameter 9.5 mm, length 4.5 m, shear stress = 4.5 N/m² lasting for 3.5 s) were compared at a flow of 5 L/min and a pressure drop of 150 mm Hg. Experiments (n = 10) with bovine blood were performed in two parallel setups using Bio-Medicus pumps BP80. Free hemoglobin in plasma (fHb) and thromboxane B₂ (TXB₂) were measured. After 6 h, the fHb increase was 31.9 ± 19.1 mg% for the throttle setup and 32.3 ± 16.2 for the tubing setup. The TXB₂ release was 296 ± 70 and 305 ± 54 pg/0.1 ml respectively after 4 h. In summary, no significant differences between the two setups for either fHb or TXB₂ could be detected. So the use of a throttle, which requires far less priming volume and a smaller blood-contacting surface while also offering a wider range of adjustment, seems preferable. Furthermore, in contrast to some other in vitro studies, these results strongly indicate that blood is quite insensitive to high shear stress exposure in the millisecond range even if repetitively applied.     

Percutaneous cardiopulmonary support with heparin-coated circuits in postcardiotomy cardiogenic shock. Efficacy and comparison with left heart bypass.

OBJECTIVE: Percutaneous cardiopulmonary support, a simplified form of venoarterial bypass, using totally heparin-coated circuits, has recently come into clinical use. To clarify its efficacy in postcardiotomy cardiogenic shock to aid weaning from cardiopulmonary bypass, we compared results of percutaneous cardiopulmonary support with those of left heart bypass using a centrifugal pump. METHODS: We reviewed 18 patients treated between 1991 and 1998 who could not be weaned from cardiopulmonary bypass. Nine were aided by totally heparin-coated percutaneous cardiopulmonary support (PCPS group), and 9 supported by left heart bypass using a centrifugal pump (LHB group). In both groups, activated clotting time was controlled at 150-200 seconds using minimal doses of heparin as needed. RESULTS: Weaning and survival rates were higher in the PCPS group than in the LHB group (100% vs 55.6%, and 66.7% vs 22.2%). The PCPS group had a smaller amount of blood loss and needed a smaller amount of blood components in the immediate postoperative period. One percutaneous cardiopulmonary support patient required surgical re-exploration for postoperative bleeding (11.1%), but no clinical thromboembolic event occurred in the PCPS group. In the LHB group, 5 patients underwent surgical re-exploration for postoperative bleeding (55.6%), and 2 underwent thrombus extirpation in the left ventricle (22.2%). CONCLUSIONS: Although this study was retrospective and historical backgrounds could have been involved, our data suggest that totally heparin-coated percutaneous cardiopulmonary support system appears more effective as an aid to weaning from cardiopulmonary bypass and in short-term circulatory support for patients in postcardiotomy cardiogenic shock.     

Interstitial fluid accumulation does not influence oxygen uptake in the rabbit small intestine

Crystalloid resuscitation increases interstitial fluid volume. Intestinal ischemia and impaired barrier function may contribute to the precipitation of multiple organ failure. Accordingly, the intestine was chosen as target organ to test whether interstitial oedema impairs oxygen extraction by the tissue.
The portal vein in anaesthetized rabbits was partially obstructed for 30 min along with an intravenous infusion of 0.9% saline 60–90 ml kg^-1 (oedema group, n = 7). Total water content of the small intestine increased from 3.4 ml g^-1 dry weight in control (n = 8) to 3.9 ml g^-1 in the oedema group ( P = 0.049). Small intestinal O₂ uptake was calculated from the arteriovenous O₂ content and electromagnetic flow measurements in the superior mesenteric artery. Mesenteric flow was reduced stepwise by a snare occluder around the artery. Intestinal oxygen-ation was monitored indirectly as well, by means of mesenteric venous lactate, arterial base excess and by mucosal pH (pH_i) assessed tonometrically.
The oxygen extraction ratios were similar in the oedema and control group at similar oxygen supplies. After a 45 min flow reduction to 15% of baseline mesenteric venous lactate and pH_i did not differ between the groups. pH; averaged 7.31 and fell to 6.74. Below an intestinal O₂ uptake of 2.5 ml min^-1, pH_i correlated somewhat better with O₂ uptake (r=0.66) than did arterial base excess (r=0.50).
The results indicate that acute elevation of extracellular volume to the extent in the present study, does not impede oxygen uptake in the gut.     

Development of an Intravascular Pumping Oxygenator Using a New Silicone Membrane

Abstract: A new intravascular pumping oxygenator (IVPO) was developed for intravascular gas exchange and circulatory assistance in critically ill patients with respiratory and circulatory failure. The IVPO utilizes new silicone hollow fibers (diameter. 1 mm: membrane width, 50 μm) and consists of two driving tubes for the oxygenation and pumping of circulating blood. The performance characteristics of the IVPO were studied using an experimental ex vivo model. With a mean hemoglobin concentration of 10.5 ± 2.3 g/dl, total oxygen transfer was 5.6 ± 1.5 ml/min at a blood flow of 200 ml/min and 6.3 ± 2.2 ml/min at a blood flow of 250 ml/min. Total CO₂ transfer was 3.8 ± 1.4 ml/min at a blood flow of 200 ml/min and 4.2 ± 1.6 ml/min at a blood flow of 250 ml/min during IVPO pumping. This preliminary experiment demonstrated that the IVPO has the capacity to function both as a circulatory assist pump and as an intravascular hollow fiber oxygenator.     

Surgical treatment with infarction exclusion technique and postoperative percutaneous cardiopulmonary support for a patient with ventricular septal perforation —A case report—

A 76-year-old woman was referred to our hospital because of postinfarction ventricular septal perforation (VSP). VSP occurred twelve days after acute myocardial infarction and resulted in interventricular shunt with Qp/Qs of 4.8. Because she suffered from pulmonary edema and oligouria, she underwent emergent surgical treatment after application of the intraaortic balloon pump. The infarction involved whole of right ventricle and the inferoposterior wall of left ventricle, and the location of the perforation was infero-septum with the size of 10 × 6 mm. A bovine pericardial patch was tailored in a triangular shape of approximately 7 × 7 × 5 cm. The base of the patch was sutured to the fibrous annulus of the mitral valve and medial margin of the patch was sewn to healthy septal endocardium. And the lateral side of the patch was sutured to the posterior wall of the left ventricle. Because of difficulty in weaning of the cardiopulmonary bypass, a percutaneous cardiopulmonary support system (PCPS) was inserted via femoral artery and vein with a flow support of 2.0 L/min. After 4 days support of PCPS, improvement of right ventricular function was detected by echocardiogram and PCPS was removed. While PCPS support, low dose heparin to maintain ACT level around 150 seconds was continuously administered. Any complication including bleeding and thrombosis was not remarkable. The patient was discharged on the 53th postoperative day, and is now doing well.