INTRODUCTION

From the inception of extracorporeal circulation technology, investigators have been searching for ways to establish the lowest priming volume for neonates and infants perfusion circuits. Between the years 1937 and 1939, the father of extracorporeal circulation as we know it today, Dr. John Gibbon, developed an experimental vertical revolving cylinder oxygenator for perfusion of the most available specimens at that time: "cats".

The apparatus was able to maintain the circulation of a 3-kilogram cat while its pulmonary artery was partially occluded (Figure 1). It was also able to maintain the entire circulation during periods of complete occlusion of the pulmonary artery, for as long as 20 minutes, with animal survival [1]. The priming volume of the apparatus was 90 ml, with an operating flow rate of 500 ml/minute. Such a momentous achievement opened the gates and ushered into the era of open-heart surgery by means of extracorporeal circulation.

Fig.1. Mrs Mary Gibbon and Charles Kraul, Dr. Gibbon's capable technician, performing a sterile experiment at the University of Pennsylvania.

Gibbon subsequently performed the first successful closure of an atrial septal defect on an 18 year old girl on May 6, 1953, using a new an improved vertical screen oxygenator [2]. Dr John Gibbon made what was once an impossible dream come true.

In 1950, Clark, Gollan, and Gupta [3] introduced an experimental pediatric oxygenator capable of oxygenating blood by a gas-dispersion method (the first system to use a silicone defoaming agent). It had also an integral heat exchanger, and a low priming volume adequate for infants. This was precursor to all bubble oxygenators later developed. In the early 1970's, the Travenol Company, introduced an infant membrane oxygenator. The unit measured 28 by 3.5 cm., composed of 0.75 sq./m of folded reinforced silicone rubber sheets 5 mm thick, with spacers made of plastic woven screens. The priming volume was 140 ml. The membrane surface area was 7,500 sq./cm., and the device was able to transfer 40 ml/oxygen/m2/min, at a blood flow rate of 300 ml/min. This membrane oxygenator was first designated as "STX290", by the manufacturer [4], and later re-named "Modulung". It was first used clinically by Dr. Sugg and colleagues [4a], at the University of Texas Southwestern Medical School (Figure 2), to perfuse a premature baby weighing 1.7 kilograms, who had an obstructed total anomalous pulmonary venous return; a record for perfusing an infant of that size at that time in history.

Fig. 2. Dr. Sugg checks a pediatric design whose small size he finds adequate.

However, the baby died on the post-operative day 11, from causes unrelated to the surgery. Correction of this type of defect during that era usually carried a mortality risk of 80%.

MATERIALS AND METHODS

The early generation of bubble oxygenators had large priming volumes, and produced plasma hemoglobin levels, usually in the range of 50 mg % to 100 mg %; whereas, with the second generation of membrane oxygenators e.g., flat plate types, "Lande-Edwards and G.E. Pierce Dual Lung", the plasma hemoglobin levels were in the range of 5 mg % to 10 mg %. The Travenol disposable membrane oxygenator was essentially the first commercially produced pediatric unit available for clinical use.

Extracorporeal circulation technology has made significant advances over the last half century, for routine use during surgical correction of acquired and congenital anomalies. Unfortunately, pediatric perfusion has not experienced the same success as adult perfusion in many areas. This is especially true in terms of the ratio of patient-blood volume to pump-prime volume. The main reason was that manufacturers at that time did not make a concerted effort to find ways of developing highly efficient biomaterials that would increase the surface area of the oxygenators and allow for an adequate gas-exchange within a small compact unit suitable for use with neonates and infants.

Historically, most oxygenators priming volume for neonates and infants range between 800 ml to 1000 ml [5,6,7,8,9]. However, in recent years, many pediatric cardiac centers are now attempting to microsize their circuits in order to reduce the prime volume; being aware of the inherent dangers associated with extreme hemodilution. For example, if a 3 kilogram neonate with an estimated blood volume of (85 ml/kg), is perfused with such a pump prime volume, it would result in a significant hemodilution of more than 300 percent, and it would certainly require post bypass blood transfusions.
Clinicians in the early days of cardiopulmonary bypass (CPB) had no other alternatives but to use bubble oxygenators (Figure 3) which required high prime volumes made up of crystalloid (clear primes) solutions with additives such as glucose, albumin 5%, 25%; dextran 40 and 70, hydroxyethylstarch 6 %, and plasma protein fractions [11,12,13,14] to raise oncotic pressure.

Fig. 3. Shiley S-070 bubble oxygenator for pediatric use.

Hemodilution has advantages and disadvantages. The advantages involve reduction in red cell concentration during CPB; and decreased blood viscosity during deep hypothermia and circulatory arrest [15], reduction of hematocrit in cyanotic infants [16]; improved organ perfusion, lower peripheral vascular resistance, especially at the onset of CPB, when arterial pressure falls below normal physiologic levels [16,17].

As the search for a reduction in prime volume continued, significant advances were made in the development of smaller membrane oxygenators, surgical techniques and postoperative care. This prompted an increase in the number of surgical procedures performed in neonates and small infants with more complex congenital heart lesions. Many of these children presented for surgical correction in their first weeks of life [5,7,8], and some even within a few days. Most of them are within the weight range of 3 to 5 kilograms, with a circulating blood volume lower than the volume required to prime the cardiopulmonary bypass circuit [6,7,8]. In 1990, Groom and colleagues [10] reported on a survey of 127 hospitals performing neonatal, infant and pediatric open-heart surgery in the United States. A total of 11,417 cases were reported. Patient's ages ranged from one day to 18 years; 15.4 % were below one month of age, 29 % were between one month and one year, and 33.8 % were between one and four years old.

The disadvantages of extreme hemodilution have been reported by various studies which demonstrate that when small infants are overdiluted, several serious side effects result, including changes in the coagulation cascade with the potential for increased bleeding in the early post-operative period [18]. In addition, there is a significant reduction in oncotic pressure [19], of as much as 50 %, due to an influx of water from the intravascular space, and an influx of albumin from peripheral stores with massive extracellular fluid shifts, and interstitial fluid accumulation [20]. In addition, hematocrit is decreased which reduces the blood's oxygen-carrying capacity [21,22]. This may also produce a hyperdynamic state after weaning from CPB with elevated cardiac output, decreased vascular resistance, and the need for excessive fluid replacement [23], pulmonary edema [20], release of stress hormones that activates the complement system [26,27], and immunosuppression with increased risks of infection [26.a]. There is also a reduction in levels of complement IgM, IgA, and IgG fractions, and in the capacity of opsonization [27]. Serum catecholamines may also be reduced by hemodilution during CPB [28].

It has also been reported that extreme and even moderate hemodilution with crystalloid components results in hyperglycemia, and osmolar flux [29], iatrogenic myocardial edema and reduced left ventricular compliance, performance and perfusion [30]. For example, if a neonate under one month of age is perfused with a pump prime volume of 650 ml to 750 ml, including whole blood to achieve a hematocrit of 20%, it results in a reduction of coagulation factor levels by 50% and decreased platelet counts to 70% of initial levels [31]. Additionally, the coagulation profile is further altered as a result of hepatic underdevelopment which continues during the first two weeks of life, albeit, with decreases in coagulation factor concentrations. Antithrombin III and fibrinogen levels are roughly 50% less than the concentration levels found in adults [32,33]. A review of the literature has shown that extreme hemodilution play an enormous part in the reduction of magnesium, calcium and albumin concentrations that continue for days post bypass [34]. These negative effects are not altered by the addition of packed red cells (to obtain a hematocrit of 35%) nor by the increased osmolarity by the addition of mannitol [30].

Plasma proteins are diluted resulting in a significant drop in their buffering capacity [36]. Dilution of clotting factors results in prolongation of the ACT values, and inadequate heparin-ization, leading to an increase in coagulation activity and a tendency to promote a bleeding diathesis post bypass [37]. Moreover, it increases capillary leakage, resulting in tissue edema, organ dysfunction, increases in pulmonary, cardiac and central nervous system morbidity [38].

To counteract the adverse effects of extreme hemodilution in the neonate or infant, we have historically added whole blood, packed red cells to the pump prime, and infuse platelets, fresh frozen plasma and cryoprecipitate at the end of CPB [39,40]. However, this introduces an additional risk and problems that have been well documented, including exposure to multiple blood-borne diseases; i.e., human acquired immunodeficiency (HIV), homologous blood syndrome CMV, hepatitis B and C, human T cell leukemia, and other infectious diseases, or a source of viral infection [26a,41,44]. There is an additional risk with the problem of transfusion reactions that may manifest immediately, or a later date from hemolytic responses. Electrolyte and substrates loads of sodium, potassium and lactate that are in stored blood can be detrimental to these very small patients [39,40].

LOW PRIME CPB CIRCUITS

Substantial efforts have been made over several years to address the issue of pump priming volumes for neonate and infants, to minimize or prevent the morbid effects of extreme hemodilution. These have been previously described in the literature [9,45,46,48,49]. We will address the microcircuits and low prime volumes used in two different cardiac centers. Both centers first reduced the length and diameter of tubing in the extracorporeal circuits, and eliminated the high prime arterial-line filters; they repositioned the entire extracorporeal circuit corresponding with the level of the patient's right atrium, and employed either vacuum assisted venous drainage or gravity venous drainage; thereby, substantially reducing the length of the venous line. The use of 3/16 inch venous and arterial lines are very important factors; with the addition of a Pall Biomedical pediatric arterial line filter, or from some low prime filter from another manufacturer. However, the current use of kinetic or vacuum assisted venous drainage is somewhat controversial.

Some cardiac centers prefer gravity drainage because it is simpler and does not require additional equipment to an already crowded perfusion circuit. It has been reported that both, kinetic-assisted and vacuum-assisted venous drainage, can produce air in the venous reservoir, with gaseous microemboli escaping into the arterial line. Rider, et al. [46], in 1998, during in vitro testing, reported on the increased transmission of air, gaseous and microemboli distal to the arterial filter which passes into the arterial line with the use of assisted venous drainage, and more so with the use of kinetic assisted venous drainage. There is also a negative pressure associated venous drainage that may increase the incidence of hemolysis in the neonates group, subsequently resulting in the need of an increase in transfusion requirements post-operatively.

CPB MICROCIRCUIT (1)

Prime Composition
* Plasmalyte-A 200 ml
* 500 IU heparin
* Sodium bicarbonate
* A Cobe Micro Neonate hollow fiber oxygenator/heat exchanger
* Venous reservoir with 3/16 in. venous and arterial tubing
* Pediatric arterial filter (Pall Biomedical)
* Arterial in-line blood gas/ temperature probe/oxygen saturation probe

Cardioplegia delivery, via mini-pump (use for dialysis)
1:1 blood cardioplegia solution via 1/8 in. tubing lines and heat exchanger

The entire circuit is placed at the same height as the operating table, and venous drainage is achieved by suction from the patient; two cardiotomy suction lines (3/16 in.) are connected to the reservoir, and controlled by a valve placed in the high-vacuum system line. The valve allows for the maintenance of a stable negative pressure (-15 to _30 mm Hg). Safety devices include an arterial line high pressure shut-off, and an air detection device placed at the outlet of the venous reservoir, which shuts off the arterial pump and clamps the arterial line if the venous reservoir level falls below a pre-set volume. Total priming volume for this system is approximately 230-250 ml, which is considerably lower than most of the (high prime) so called microsize circuits that are currently in clinical use (500-600 ml). In addition, a hemoconcentrator is deployed for modified ultrafiltration (MUF) if there is any residual perfusate in the system.

Conventional ultrafiltration using "Amicon" filters during CPB, has been in practice since the early 1980's. The MUF procedure was developed to decrease the detrimental effects of elevated total body water post bypass [47], which results in what is known as the "capillary leak" syndrome, which produces an increase in tissue fluid, and contributes to mechanisms ranging from direct physical consequences of extreme hemodilution to fluid overload and an exacerbated inflammatory response [47]. This syndrome is also associated with the systemic inflammatory response that occurs after CPB, especially in neonates, infants and children [51,52]. The inflammatory mediator response associated with CPB remains a major cause of morbidity and mortality post-bypass, especially in infants and children [47,51]. MUF can be deployed at any time during and post-bypass. Presently, in the United States, only "Minntech Corp." supplies hemoconcentrators for neonates and small infants. It has also been reported that there is another positive aspect from the use of hemoconcentrators. It allows for several of the inflammatory mediators that are "small" enough in molecular weight to cross the highly permeable membrane of the hemoconcentrator [50,51,52, 53].

There are other studies which indicate that MUF improves cerebral metabolic recovery after circulatory arrest [55,53]. Jansen et al. [54] reported that the reduction in prime volume attenuates the hyperdynamic response after CPB in neonates and children. Bando, et al. [56] also reported on the benefits of MUF in high-risk patients undergoing operations for complex congenital heart disease.

CPB MICROCIRCUIT (2)

Cardiac center (2) has developed their microcircuit by decreasing the diameter of the tubing instead of the lengths, and employs gravity venous drainage for several reasons. They have indicated that venous gravity is simpler and does not require additional equipment to the already hardware-laden heart lung machine. Their system is safe and simple, and can be used routinely for all neonates, with a priming volume low enough to maintain a post dilutional hematocrit of 25 percent with minimal or no transfusion requirement.

Cobe-Micro Hollow-fiber Oxygenator
Arterial line.................1/8 in I.D.
Venous line ............... 3/16 in. I.D.
Pump tubing raceway.....3/16 in. I.D.
A short segment of 1/4 in. I.D. tubing is connected to the venous line as it enters the reservoir.
Cardiotomy suction lines 3/16 in. I.D.; all lines are made as short as possible.
Oxygen saturation and hematocrit electrodes are inserted in the venous line.
A low level arterial pump sensor and bubble detector that stops the arterial pump are also mounted.
Total priming volume is approximately 160 ml, for flow rates calculated up to 700 ml/min. In addition, a low prime hemoconcentrator is inserted in the circuit, which can process as little as 15 ml of residual perfusate left in the extracorporeal circuit. Diuresis is aggressively managed followed post-bypass.

PRIME COMPOSITION

Plasmalyte
Albumin 25 % 50 ml
Sodium bicarbonate 5 mEq or 1/2 mEq/kg
Heparin 500 IU
Manitol 250 mg/kg
Aprotinin 4.6 mg/kg
Calcium chloride 50 mg/unit of fresh donor blood if needed
If the calculated hematocrit is less than 20 % to 25 %, packed red blood cells are added.

CARDIOPLEGIA

1:1 blood cardioplegia delivery via Sorin BCD system, primed with 70 ml of blood to which the patient is not exposed, and is flushed to the Cell Saver.

CONCLUSION

We have illustrated two systems among many that are attempting to address the age old problem of extreme hemodilution in the neonate and infant group, presented for extracorporeal circulation. However, we must bear in mind that they have not reached the ultimate goal in resolving this problem. The experience gained through these first steps, show the technical feasibility of eliminating extreme hemodilution, and conducting a safe and uneventful perfusion without the need for post-bypass transfusions.

Darling and colleagues [57] remind us that there are major potential problems with micro-sizing extracorporeal circuits, especially those with vacuum assisted venous drainage. Some of them have been identified as follows:

* Inadequate venous reservoir venting which is potentially dangerous due to over-pressurization of the reservoir with an integrated cardiotomy, from the suction lines.
* Increased negative pressure that can occur, resulting in an in increase in hemolysis.
* The potential for increased shunt flows.
* Slower reaction time due to the relatively small volume of perfusate maintained in the venous reservoir.
* If an arterial roller pump is used in the circuit, the possibility of under-occlusion exposes the oxygenator to a vacuum pressure, that could lead to air been drawn across the membrane fibers.

The ideal circuit for neonatal and small infant cardiopulmonary bypass still needs to be designed and built. Some improvements on the circuits used in our days have allowed for a reduced prime and possibly led to a decreased inflammatory response.

REFERENCES

1. Gibbon, J. H., Jr.: The maintenance of life during experimental occlusion of the pulmonary artery followed by survival. Surg. Gynec. Obstet. 69: 602, 1939.

2. Gibbon, J. H., Jr., Miller,B.J., Fineberg, C.: An improved mechanical heart lung apparatus. Med.Clin.N.Amer. 37: 1603,1953.

3. Clark, L.C., Gollan, F., and Gupta, V.B.: The oxygenation of blood by gas dispersion. Science 111: 85, 1950.

4. Douglas, M., Brinbaum,D., and Eiseman,B.: Biological Evaluation of a Disposable Membrane Oxygenator. Arch Surg. 103;89-92,July, 1971.

4a. Medical World News/April 6, 1973

5. Cappola, P.R.,et al.: Cardiopulmonary Bypass in the Neonate: Case Report. J Extra-Corp Tech. Volume.13 No 6, 1981.

7. Hartley-Winkler,M.,et al.: Perfusion Considerations For Infants 10 kilograms or less. J.E.C.T (17:31-36, 1985.

8. Castenada, A.R., Lamberti,J., Sade, R.M., Williams, R.G., Nadas,A.S.: Open-heart surgery during the first three months of life. J Thorac Cardiovasc Surg.68:719,1974.

9. Sistino, J.J., Mongero, L.B., Michler, R.E., et al.: Dilutional effects of cardiopulmonary bypass in patients less than 15 kilograms using low-prime circuits. Proc Am Acad Cardiovasc Perf. 1993;14:98-103.

10. Groom, R.C., Hill, A. G., Aki, B.F., Lefrak, E.A., and Krusz, M.: Pediatric perfusion survey . Proceed Am Acad Cardiovasc Perfu- sion 1990;11:78-84.

11. Panico, F.G., Neptune, W.B.: A mechanism for eliminating donor blood from the pump oxygenator. Surg Forum 1959;10:605-609.

12. Zudhi, N., McCollough, B., Carey, J., Greer, A.: A double helical reservoir heart lung machine designed for hypothermic perfusion primed with five percent glucose in water inducing hemodilution. Arch Surg 1961;82:320

13. Laks, H., et al: The effects of Cardiopulmonary Bypass with Crys- talloid and Colloid Hemodilution on Myocardial Extravascular water J.Thorac Cardiovasc Surg. 73; 129,1977.

14. Laver, M.B. Buckley, M.J. Austen, W.G., et al.: Extreme hemodilu- tion with profound hypothermia and circulatory arrest. Bibl Haema- tol. 1975;41:225-238.

15. Leveen, H.H.,et al.: Lowering viscosity to overcome vascular resis- tance . Surg Gynecol Obstet 1980; 139-49.

16. Gordon, R. J., Ravin, M., Rawitscher, G.R. : Changes in arterial pres sure, viscosity and resistance during cardiopulmonary bypass. J Tho- rac Cardiovasc Surg. 1975; 69:552-561.

17. Kalter, R.D., et al.: Cardiopulmonary bypass associated hemostatic abnormalities. J Thorac Cardiovasc Surg. 1997;77:427-435.

18. Huyzen,R.J., van Oeveeren,W., Wei,F.,et al.: In vivo effect of hemodilution on Act and high-dose thrombin time during cardiopul- monary bypass. Ann Thorac Surg. 1996;83:459-65.

19. Shupbach, P., Perfusate oncotic pressure during cardiopulmonary bypass, optinum levels as determined by metabolic acidosis, tissue edema, and renal function. Vox Sang. 35:332-334,1978.

20. Puri, V.K., Michaels, S., and Carlson,R. W.: Pulmonary edema asso- ciated with reduction in Plasma oncotic pressure.Surg Gynecol. Obstet. 151: 341-48, October, 1980.

21. Marty, A.T., Prather, J., Matloff, J.M., and Schauble, J.: Oncotic effects of dilutional bypass, albumin and diuretics. Arch Surg. 107: 21, July 1971.

22. Beattie, H.W., Evans, G., Garnet, E.S., et al.: Albumin and water fluxes during cardiopulmonary bypass. J Cardiovasc Surg. 1974; 67: 926-931.

23. Wiley, J.,et al.: Effects of hypothermia, hemodilution and pump oxy- genation on water content, blood flow, and oxygen delivery and renal function. 1981; 31(2): 121-123.

24. Cohn, L.H., Angell, W.A., Shumway, N.E., et al.:Body fluid shifts after cardiopulmonary bypass: Effects of congestive heart failure and hemodilution. J Thor Cardiovasc Surg. 1971: 62:423-30.

25. van Velzen,-Bald, H., Dijkstra, Y., Schurink, G.A., et al.: Cardio- pulmonary bypass and host defense functions in human beings: 1. Serum levels and role of immunoglobins and complement in phago- cytosis. Ann Thorac Surg. 1985;39: 207.

26. Anand, K.J.,, Hensen, D.D., Hickey, P.R.: Hormonal metabolic stress responses in neonates undergoing cardiac surgery. Anesthesiology 1990;73;661-670.

26a. Frater,R.M.W., Santos,G.H.: Sources of infection in open-heart sur- gery. N.Y. State J Med 1974;74: 2368-74.

27. Collet, B.,et al.: Pathways to complement activation during cardio- pulmonary bypass. Br. Med J 1984;1251-54.

28. Balassaraswathi, K., et al.: Effects of volume on serum catechol- amines during cardiopulmonary bypass. Can Anesth Soc J. 1980; 27:135-39

29. Benzing, G, 111, et al.; Glucose and insulin changes in infants and children undergoing hypothermic open-heart surgery. Am J Car- diol. 1983; 52: 133-36.

30. Foglia,R.P., Lazar, H.L., Steed, D.L., et al.: Iatrogenic myocardial edema with crystalloid primes: Effects on left ventricular compliance performance and perfusion. Surg. Forum. 1978; 29: 312-15.

31. Kern, F.H., Morana, N.J., Sears, J.J., et al.: Coagulation defects in neonates during cardiopulmonary bypass. 1992;54:541-46.

32. Andrew,M., Paes, B., Milner, R.E., et al.: Development of the human coagulation system in full term infant. Blood. 1987: 165-72.

33. Morselli, P.L.,Franco-Morselli,R., Bossi, L.,et al.: Clinical pharmo- kinetics in newborn infants. Clin Pharmokinet 1980;5:485-91.

34. Adams, A., Burns, J.E.: The biochemical consequences of cardio- pulmonary bypass in children. Clinica Chimica Acta. 1979; 93: 101-11.

35. Foglia, R.P., Lazar, H.L., Steed,D.L.,et al.: Iatrogenic myocardial edema with crystalloid primes: Effects on left ventricular compli ance, performance, and perfusion. Surg Forum. 1978;:29312-15.

36. Reeves,R. B., et al.: An imidazole alphastat hypothesis for veterbrate acid-base regulation: Tissue carbon dioxide content and body tem- perature in bullfrogs.Resp. Phys. 1972; 14: 219-36.

37. Huyzen, R.J., van Oeveren, W., Wei, F., et al.: In-vitro effect of hemodilution on ACT and high-dose thrombin time during cardio- pulmonary bypass. Ann Thorac Surg. 1996;62:533-37.

38. Ashraf, S.S., Tian, Y., Zacharrias, S., et al.: Effects of cardiopulmon- ary bypass on neonatal and pediatric inflammatory profiles. Eur J Cardio-Thorac Surg. 1997; 12: 862-68.

39. Mano, C.S., Hedberg, K.W., Comparison of the hemostatic effects of whole blood, stored whole blood, and components after open-heart surgery in children. Blood. 1991;77: 930-36.

40. Ratcliffe, J.M., Elliott, M.J., Wyse, R. K., et al.: The metabolic load of stored blood: Implications for major transfusions in infants. Arch Dis in Childhood. 1986;61: 1208-14.

41. Gadboys, H. L., Slonim, R., Litwak, R.S., Homologous blood syn- drome : 1. Preliminary observations on its relationship to clinical cardiopulmonary bypass. Ann Surg. 1962; 156: 793-804.

42. Ratcliffe, J,M, Hunter, S., et al.: The role of priming fluid in meta- bolic response to cardiopulmonary bypass in children 15 kilograms of body weight undergoing open-heart surgery. J Thorac Cardiovasc Surg. 1988; 36: 2; 64-74.

43. Lilleasen,P., Stokk, S. Q.: Moderate and extreme hemodilution in open-heart surgery: Fluid balance and acid-base studies. Ann Throac Surg. 1978;25:127-33

44. Engleman, R.M., Chase, R.M. Jr., Boyd, A.D., Reed, G.E.: Lethal postoperative infections following cardiac surgery: review of four years experience. 1973; 48(suppl111J 390-393.

45. Turina, M., Housman, L.B., Intaglietta, M., et al.: An automatic car- diopulmonary bypass unit for use in infants. Trans Amer Soc Artif Int Organs. 1971;27:376-80.

46. Rider, S.P., Simon, L. V., Rice, B.J., et al.: Assisted venous drain- age, venous air, and gaseous microemboli transmission into the arte- rial line: An in vitro study. J Extra-Corp. Technol. 1998;30;160-65.

47. Elliott, M.J., et al.: Ultrafiltration and modified ultrafiltration in pediatric open-heart operations. Ann Thorac Surg. Dec; 56(6): 1518-1521.

48. Housman, L.B., et al.: Use of hemodilution during total cardiopmon- ary bypass in neonates: Experimental evaluation employing a mini- aturized heart-lung machine. Surgery. 1972;72: 460-65.

49. Wabeke, E., et al.: Clear prime for cardiopulmonary bypass. A mini- aturized circiut. J Cardiovasc Surg. 1988;29:117-22.

50. Groom, R., Bechara, A., Albus, R., et al.: Alternative method of ultrafiltration after cardiopulmonary bypass. Ann Thorac Surg. 1994 Aug,58(2):573-74.

51. Journois, D., Pouard, P., Greeley, W.J.,et al.: Hemofiltration during cardiopulmonary bypass in pediatric cardiac surgery: effects on hemostasis, cytokines, and complement components. Anesthesiology 1994 Nov;81(5):1181-9; discussion. 26a-27A.

52. Milar, A., Armstrong, L., et al.: Cytokine production and hemofil- tration in children undergoing cardiopulmonary bypass. Ann Thorac Surg 1993 Dec, 56(6): 1499-502.

53. Andreasson, S., Gothberg, S., et al.: Hemofiltration modifies com- plement activation circulation in infants. Ann Thorac Surg, Dec,56(6):1515-7.

54. Jansen, P.G., et al.: Reduction in prime volume attenuates the hperdynamic response after cardiopulmonary bypass. Ann Thorac Surg. 1995;60: 544-49.

55. Skaryak, L.,et al.: Modified ultrafiltration improves cerebral meta bolic recovery after circulatory arrest. J Thorac Cardiovasc Surg. 1995 Apr,109(4): 744-51.

56. Bando, K.,et al.: Benefits of modified ultrafiltration in high-risk patients undergoing operations for complex congenital heart disease. Presented at the 34th Annual Meeting of The Society of Thoracic Sur- geons 1998 Jan 26-28, New Orleans, LA.

57. Darling, E., Kaemmer,D., Lawson, S., et al.: Experimental use of an ultralow-prime neonatal cardiopulmonary bypass circiut utilizing as sisted-venous drainage. J Extra-Corp Technol. 1998;30:184-89.