Blood bank transfusion is necessary and lifesaving in pediatric open-heart surgery; however, along with its benefits, this form of blood therapy also carries significant risks. Furthermore, blood transfusion is often poorly accepted by the parents. With modern blood banking procedures, AIDS and hepatitis C are no longer considered major transfusion-related risks.1 Today, immunomodulation and inflammation are the main concerns following allogeneic blood transfusion, as they significantly contribute to cardiopulmonary dysfunction, organ failure, and sepsis after surgery. As a result, there is an increase in the length of ICU and hospital stays and subsequently in hospital costs.1–4 The efficiency of a downsized, low prime volume circuit in decreasing blood transfusions and in increasing the incidence of bloodless surgery has been previously demonstrated.3–6 The aim of this work is to assess the potential benefit of vacuum-assisted venous drainage (VAVD) for further reducing prime volume and blood transfusion.
Materials and Methods
Approval to conduct this retrospective anonymous study without the need for parental consent was obtained from our institutional review board.
All patients weighing less than 10 kg who underwent surgery between February 2007 and December 2007 were considered in this study, including those who underwent reoperation.
The cardiopulmonary bypass (CPB) system consisted of the Kids D100 hollow-fiber membrane oxygenator (Sorin-group, Mirandola, Italy) connected to 3/16 in. diameter arterial and venous silicone tubing lines. The oxygenator hard shell reservoir has a built-in valve that opens when reservoir pressure is above 5 mm Hg or under −80 mm Hg. To limit the prime volume, we do not add an arterial filter or a hemofilter. The total priming volume of the circuit was 120 ml. The priming compound was either blood, fresh frozen plasma and packed red blood cells, or a mixture of 50% lactated Ringer’s solution and 50% hydroxyethyl starch. Our strategy was to maintain a hemoglobin level of at least 8 g/dl during the procedure, so the choice of prime composition depended on the predicted hemoglobin level during bypass. This estimated level was calculated according to an empirical formula:
Predicted hemoglobin (g/dl) = patient blood volume (ml) × preoperative hemoglobin (g/dl)/patient blood volume (ml) + prime volume (ml).
The patient blood volume valuation was 80 ml/kg.
Blood drainage was optimized with 0.5 mm wall thickness metallic venous cannulae with a terminal side hole (Soframedica, Reventin-Vaugris, France). The venous return was gravity siphon-dependent and vacuum-assisted: the height differential between the patient and the blood level in the oxygenator reservoir was around 10 cm and the hard shell reservoir negative pressure was between −20 and −30 mm Hg. Negative pressure was applied through a vacuum regulator (Precision Medical, Northampton, PA) and continuously monitored with a pressure transducer displayed on the heart-lung machine screen (Stoëckert S5, Munchen, Germany). Patients were operated on using a uniform anesthetic technique. Anesthesia was induced with sufentanyl 0.5 μg/kg, flunitrazepam 0.03 mg/kg, and pancuronium 0.1 μg/kg dosages adjusted to individual susceptibilities. After tracheal intubation, continuous infusion of sufentanyl (1–2 μg/kg/h) was associated with bolus injection of flunitrazepam (0.03 mg/kg) every hour. For patients weighing more than 9 kg, propofol was substituted for flunitrazepam. The initial dosage of propofol was 1–2 mg/kg followed by a continuous injection of 5–10 mg/kg/h. Normothermic perfusion (37°C) and intermittent warm blood cardioplegia were used for all the patients, with techniques that have been previously described.7 Prime was heated to 37°C before bypass circuit connection to the patient, and the water heater was set to 37.5°C during CPB time. The cardioplegic solution added to arterial blood is marketed under the name “CP1B solution” by Pharmacie Centrale des Hôpitaux de Paris, Assistance Publique and is composed of 0.8 mmol/ml potassium, 0.8 mmol/ml magnesium, 2.45 mmol/L chloride, and 0.05 mmol/ml procaine. During aortic cross-clamping, after the initial cardioplegia, reinjections were performed every 20 minutes.
Blood flow was maintained during bypass at 2.7 L/min/m2 with a tolerance of 15% fluctuation. Before bypass was completed, patients at risk of postoperative low cardiac output syndrome were injected with a 1-mg/kg loading dose of enoximone during a 10-minute period followed by a continuous infusion at 10 μg/kg/min. Eligible patients for enoximone infusion were selected on the criteria used in the PRIMACORP studies (Table 1).8,9 Epinephrine use was limited to patients with low arterial pressure. After bypass, the circuit blood was collected or, when intraoperative transfusion was required, pooled with residual bank blood products. In the group with asanguineous prime, this blood was systematically reinfused into the patient. In the group with blood prime, this blood was kept at 4°C for 6 hours and used whenever necessary. In most cases, diuresis and hemoconcentration were supported by furosemide. The adequacy of perfusion was assessed based on serial serum lactate level measurements. The time to extubation was considered a global estimate of outcome quality.
This study included 150 patients. The patients’ characteristics, CPB durations, aortic cross-clamping times, and diagnoses are presented in Tables 2 and 3. Their body surface area varied from 0.18 to 0.46 m2 and, therefore, their theoretical bypass flows varied from 0.5 to 1.2 L/min. Venous return through 3/16 inner diameter tubing was achieved without specific problems.
The incidence of blood transfusion is detailed in Table 4. It is noteworthy that peri- and postoperative blood product transfusions were limited for every patient to 1 unit of packed red blood cells and 1 unit of fresh frozen plasma (except for one patient who had 2 units of blood cells and one patient who did not receive plasma). None of the patients underwent platelet infusion. Consequently, the number of donors to which a patient was exposed was two in 98% of the transfused patients throughout their operative time and intensive care stay.
The variations in hemoglobin levels are presented in Table 5. In blood-free patients, the median postoperative hemoglobin level was 10.6 g/dl and the nadir median hemoglobin level of 8.7 g/dl was reached during the procedure. Pre-, peri-, and postoperative serum lactate levels and the times to extubation are presented in Table 6.
There is increasing concern about the safety of blood transfusion. Postoperative hemodynamic instability, pneumonia, and transfusion-related acute lung injury are clearly linked to the number and/or age of blood units transfused.3,4,10,11 Transfusion-related acute lung injury is an under-recognized complication, but it is, nevertheless, the most important cause of transfusion-related death in United States.10 In adult surgery, a 5% increased risk of pneumonia for each unit of packed red blood cells transfused has been estimated.10 Furthermore, transfusion is associated with prolonged length of stay in ICU and in hospitals.2,3,10 Currently, there is no consensus on the “critical hemoglobin level” for pediatric patients.12 This is most likely due to the numerous factors that affect the critical hemoglobin level, including patient age, cardiopathy, and preoperative hemoglobin level. It is likely that this critical hemoglobin level varies during anesthesia, CPB, and after surgery in an active patient. As a result, the criterion upon which transfusion is decided is still up for debate.13 In the literature, large differences in permissible hematocrit levels are reported. Some authors consider a predicted hematocrit value upon CPB of at least 16% to be acceptable,14 whereas others consider a value under 20% as a transfusion criterion,15 and yet others stress the risk associated with a hematocrit level under 25%.12 We arbitrarily chose 8 g/dl of hemoglobin, equivalent to a hematocrit level of 25%, as a tolerable nadir hemoglobin level during bypass. This level is one of the highest shown in the literature. Furthermore, in a previous study, we experienced a rapid post-bypass increase in hemoglobin levels after transfusion of circuit blood, with or without lasix injection.6 This increase is confirmed in the present work: the median hemoglobin level of 8.7 g/dl during bypass rose to 10.6 g/dl in intensive care.
Bloodless surgery patients had a satisfactory clinical tolerance of hemodilution as assessed with normal serum lactate levels during and after bypass as well as with the short time to extubation. In this study, we never experienced a negative side effect because of low hemoglobin levels, probably because they were preemptively treated by allogeneic blood transfusion.
It was recently demonstrated that decreasing the prime volume is an efficient way to decrease blood transfusion.3,5,6 From a perfusionist’s point of view, simplifying the bypass circuitry by downsizing its components is the best way to decrease prime volume. The use of VAVD decreases the need for gravity-dependent blood drainage and allows the arterial pump and the membrane oxygenator to be heightened (Figure 1), thus decreasing the length of the venous and arterial lines. This decrease in line length not only decreases the prime volume, but also decreases the resistance to venous blood drainage. We have uniformly decreased the prime volume to 120 ml for <10 kg patients, i.e., for arterial flow up to 1200 ml/min. In our experience, using 3/16 in. venous lines and conventional gravity venous drainage resulted in maximal CPB flow of 850 ml/min for a maximal patients’ weight of 6 kg.6 With VAVD, the maximal flow through 3/16 in. tubing is increased by 40%. This is the major reason for the improved percentage of transfusion-free operations. The reduction in blood prime volume obtained by shortening of 3/16 tubing is limited. The volume of 1 m of 3/16 tubing is 17.8 ml, which is approximately the gain we obtained by introducing VAVD. This small reduction in prime volume could hardly explain a decrease of blood transfusion. The better drainage obtained with VAVD is probably related to the other effect of line shortening, which is decreased resistance. With conventional gravity venous drainage, all patients weighing <6 kg were transfused and the cut off point for bloodless prime was 8 kg. At that time, there were only eight patients out of 44 (18%) weighing between 6 and 8 kg operated on with blood-free transfusion surgery.6 With VAVD this percentage was increased to 27% (7 of 27 patients).
There is no consensus regarding the best way to assess perfusion adequacy, and there is no perfect marker of CPB optimal quality. Oxygen-derived parameters like mixed venous saturation, is a classical marker that is easy to monitor online. However, the occurrence of regional deoxygenation despite normal mixed venous saturation has been described.16,17 The optimal levels for cyanotic and acyanotic patients are probably different, and the correlation between venous saturation and anaerobic energy supply is poor. Consequently, intermittent whole blood lactate measurements have been proposed as an index of perfusion adequacy.18 Blood lactate is better for detecting the correct match of oxygen supply and demand than mixed venous oxygen saturation.19 We choose sequential blood lactate levels as a marker of hypoperfusion. The time to extubation was used as a criterion of hemodilution tolerance, ill-tolerated acute anemia being most unlikely to be compatible with easy and fast weaning from mechanical ventilation.
Negative effects of VAVD have been previously described, including an overestimation of arterial blood flow and an increase in gaseous microemboli.20,21 Most of these negative effects are avoidable with good practice using this new technique. A high vacuum level (−65 mm Hg) is associated with an increase in gaseous microemboli downstream of the arterial filter, whereas the level of embolic activity observed with a lower vacuum level (−40 mm Hg) is equivalent to that seen with gravity siphon-dependent venous drainage.22 The efficacy of VAVD in decreasing blood transfusion has been described in pediatric patients with the use of negative pressure between −10 and −40 mm Hg and has not been associated with specific complications.19 Whatever the method of venous drainage used, the major factor that induces gaseous microemboli in the arterial line is the presence of air in the venous line.23
We believe that the “total driving force” equivalent to the sum of gravity-dependent venous drainage (in cm) and VAVD (in mm Hg) must remain at a reasonable level, i.e., a level equivalent to 60–70 cm of gravity-dependent venous drainage. The ultimate goal is to replace the force of gravity by that of vacuum and to raise the rotor pump and membrane oxygenator to the patient level. We are currently developing this model with ongoing laboratory experimentation.
We consider VAVD to be an effective tool to significantly decrease blood transfusion and with good practice, the drawbacks associated with this technique are avoidable. However, we do believe that the beneficial effects on the clinical outcome presented here should be confirmed by further studies.
1. Spiess BD: Blood transfusion: The silent epidemic. Ann Thorac Surg
72: S1832–S1837, 2001.
2. Murphy GJ, Angelini GD: Indications for blood transfusion in cardiac surgery. Ann Thorac Surg
82: 2323–2334, 2006.
3. Karamlou T, Schultz JM, Silliman C, et al
: Using a miniaturized circuit and an asanguineous prime to reduce neutrophil mediated organ dysfunction following infant cardiopulmonary bypass. Ann Thorac Surg
80: 6–14, 2005.
4. Church GD, Price C, Sanchez R, Looney MR: Transfusion-related acute lung injury in the paediatric patient: Two case reports and a review of the literature. Transfus Med
16: 343–348, 2006.
5. Golab HD, Takkenberg JJ, van Gerner-Weelink GL, et al
: Effects of cardiopulmonary bypass circuit reduction and residual volume salvage on allogeneic transfusion requirements in infants undergoing cardiac surgery. Interact Cardiovasc Thorac Surg
6: 335–339, 2007.
6. Durandy Y: Usefulness of low prime perfusion pediatric circuit in decreasing blood transfusion. ASAIO J
53: 659–661, 2007.
7. Durandy Y, Hulin S: Normothermic bypass in pediatric surgery: Technical aspect and clinical experience about 1400 cases. ASAIO J
52: 539–542, 2006.
8. Hoffman TM, Wernovsky G, Atz AM, et al
: Prophylactic intravenous use of milrinone after cardiac operation in pediatrics (PRIMACORP) study. Prophylactic intravenous use of milrinone after cardiac operation in pediatrics. Am Heart J
143: 15–21, 2002.
9. Hoffman TM, Wernovsky G, Atz AM, et al
: Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease. Circulation
107: 996–1002, 2003.
10. Vamvakas EC, Carven JH: Transfusion and post operative pneumonia in coronary artery bypass graft surgery; effect of length of storage of transfused red cells. Transfusion
39: 701–710, 1999.
11. Kleinman S, Caulfield T, Chan P, et al
: Toward an understanding of transfusion-related acute lung injury: Statement of a consensus panel. Transfusion
44: 1774–1789, 2004.
12. Jonas RA, Wypij D, Roth SJ, et al
: The influence of hemodilution on outcome after hypothermic cardiopulmonary bypass: Results of a randomized trial in infants. J Thorac Cardiovasc Surg
126: 1765–1774, 2003.
13. Ferraris VA, Ferraris SP, Saha SP, et al
: Perioperative blood transfusion and blood conservation in cardiac surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists clinical practice guideline. Ann Thorac Surg
83: S27–S86, 2007.
14. Van Son JAM, Hovaguimian H, Rao IM, et al
: Strategies for repair of congenital heart defects in infants without the use of blood. Ann Thorac Surg
59: 384–388, 1995.
15. Miyaji K, Kohira S, Miyamoto T, et al
: Pediatric cardiac surgery without homologous blood transfusion using a miniaturized bypass system in infants with lower body weight. J Thorac Cardiovasc Surg
134: 284–289, 2007.
16. Schmid FX, Phillipp A, Foltan M, et al
: Adequacy of perfusion during hypothermia: Regional distribution of cardiopulmonary bypass flow, mixed venous and regional venous oxygen saturation-hypothermia and distribution of flow and oxygen. Thorac Cardiovasc Surg
51: 306–311, 2003.
17. Lindholm L, Hansdottir V, Lundqvist M, Jeppsson A: The relationship between mixed venous and regional venous oxygen saturation during cardiopulmonary bypass. Perfusion
17: 133–139, 2001.
18. De Somer F: What is optimal flow and how to validate this. J Extra Corpor Technol
39: 278–280, 2007.
19. Ranucci M, Isgro G, Romitti F, et al
: Anaerobic metabolism during cardiopulmonary bypass: Predictive value of carbon dioxide derived parameters. Ann Thorac Surg
81: 2189–2195, 2006.
20. Fiorucci A, Gerometta PS, DeVecchi M, et al
: In vitro assessment of the vacuum-assisted venous drainage (VAVD) system: Risks and benefits. Perfusion
19: 113–117, 2004.
21. Corno AF: Systemic venous drainage: Can we help Newton? Eur J Cardiothorac Surg
31: 1044–1051, 2007.
22. Jones TJ, Deal DD, Vernon JC, et al
: Does vacuum-assisted venous drainage increase gaseous microemboli during cardiopulmonary bypass? Ann Thorac Surg
74: 2132–2137, 2002.
Copyright © 2009 by the American Society for Artificial Internal Organs
23. Nakanishi K, Shichijo T, Shinkawa S, et al
: Usefulness of vacuum-assisted cardiopulmonary bypass circuit for pediatric open-heart surgery in reducing homologous blood transfusion. Eur J Cardiothorac Surg
20: 233–238, 2001.