|
|
||||||||
Ann Thorac Surg 2003;75:1506-1512
© 2003 The Society of Thoracic Surgeons
a Department of Anesthesiology, Mayo Foundation, Rochester, Minnesota, USA
b Division of Cardiovascular Surgery, Department of Surgery, Mayo Foundation, Rochester, Minnesota, USA
c Department of Biostatistics, Mayo Foundation, Rochester, Minnesota, USA
Accepted for publication December 12, 2002.
* Address reprint requests to Dr Oliver, Department of Anesthesiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905, USA
e-mail: oliver.william{at}mayo.edu
| Abstract |
|---|
|
|
|---|
METHODS: We prospectively randomized 56 patients weighing 10 kg or less who required cardiopulmonary bypass to receive either one unit of fresh-frozen plasma or 200 mL of albumin 5% in the prime. After protamine administration, samples for prothrombin time, fibrinogen, platelet count, and thromboelastogram were obtained. Mediastinal chest tube drainage and transfusion requirements were documented.
RESULTS: There were no significant differences between groups regarding demographic or surgical characteristics. Blood loss during the first 24 hours was similar in both groups, but total transfusions were significantly greater in those who received fresh-frozen plasma instead of albumin 5% in the prime (8.0 ± 4.2 versus 6.1 ± 4.5 U, respectively; p = 0.035). Post hoc analyses suggest that for cyanotic patients and patients undergoing complex operations, fresh-frozen plasma in the prime results in less blood loss than albumin 5%.
CONCLUSIONS: Substitution of albumin 5% for fresh-frozen plasma in the prime of acyanotic patients weighing 10 kg or less who undergo noncomplex operations requiring cardiopulmonary bypass significantly reduces perioperative transfusions without increasing blood loss. Further investigation is needed to determine whether increased blood loss is associated with increased transfusions when albumin 5% is substituted for fresh-frozen plasma in the prime of infants and children who are cyanotic or undergoing complex operations.
| Introduction |
|---|
|
|
|---|
Hemodilution occurs because the priming volume necessary to initiate CPB usually exceeds the infants calculated blood volume by 100% to 400%. Additional fluid is frequently required during the rewarming phase of CPB, which further exacerbates hemodilution. Consequently, the concentration of platelets, and clotting factors, particularly fibrinogen, may be reduced by 50% to 75% [11]. To counteract this, fresh whole blood has been added to the prime for infants with excellent results [8]. However, many hospitals do not provide fresh whole blood or even whole blood (>48 hours subsequent to donation) [12]; alternatively, a combination of fresh-frozen plasma (FFP) and packed red blood cells (PRBC) is administered [1]. Although viral transmission from blood has decreased significantly since the institution of donor screening programs in the 1990s [13], the overall calculated risk for hepatitis B and C, and human immunodeficiency virus for each unit of FFP administered is 1 in 34,000 [14]. The risk is low but increases dramatically as the aggregate risk of multiple exposures to FFP often occurs in patients undergoing congenital heart repair.
The aim of this study was to evaluate blood loss and transfusion requirements for infants weighing 10 kg or less with either FFP or albumin 5% added to the prime for CPB.
| Material and methods |
|---|
|
|
|---|
Patients were randomized to receive either 200 mL of albumin 5% or one unit of FFP in the prime for CPB. All personnel associated with the perioperative care of these infants and children except the perfusionists were blinded to the constitution of the prime.
Anesthetic and surgical techniques
Anesthesia consisted of 50 µg/kg of fentanyl, 0.1 mg/kg midazolam, pancuronium, and isoflurane. Routine monitoring for an infant undergoing open heart operation was used including pulse oximeter, mass spectrometer, and nasopharyngeal temperature probe. The femoral artery was cannulated for continuous blood pressure measurements and access for arterial blood gas and coagulation tests. Central venous access was obtained in the jugular vein for monitoring central venous pressure and infusing vasoactive agents. After intubation, patients were connected to a 900 Servo C (Siemens-Elema, Solna, Sweden) ventilator. Ringers lactate solution or 0.2 N saline solution was infused for fluid maintenance. Patients did not receive antifibrinolytic agents.
Anticoagulation was accomplished with 300 U/kg porcine heparin (Elkins-Sinn Inc, Cherry Hill, NJ) and monitored with an activated clotting time (ACT) performed with a Hemochron 800 (International Technidyne Co, Edison, NJ). The ACT was maintained greater than 450 seconds during CPB with additional heparin administration as required. The CPB circuit was primed with Plasmalyte, calcium chloride, sodium bicarbonate 8.5%, mannitol 12.5%, and either albumin or FFP depending on the randomization. Washed PRBC were added to the prime to maintain a hematocrit between 21% and 25% on initiation of CPB.
Cardiopulmonary bypass was performed with nonpulsatile flow of 2.2 Lmin-1m-2 and membrane oxygenator (SCI-Med, Minneapolis, MN) with priming volumes ranging between 800 and 1,200 mL. Hypothermia was induced in most patients and monitored by nasopharyngeal temperature. When circulatory arrest was used, patients received 5 mg/kg thiopental, 0.5 g/kg mannitol 12.5%, and 0.5 mg/kg dexamethasone immediately before interruption of circulation. The senior cardiac surgeon (F.J.P.) classified each operation as simple or complex.
Coagulation and transfusion
Immediately after placement of an arterial catheter, samples for arterial blood gas, ACT, activated partial thromboplastin time, prothrombin time (PT), fibrinogen, and thromboelastogram were collected. Samples were obtained after withdrawal of 10 mL of blood from the catheter and returned to the patient. Thromboelastogram measurements included reaction time, angle (
), and maximum amplitude. An ACT and arterial blood gas were repeated 5 minutes after initiation of CPB and on completion of rewarming before separation from CPB. Ten minutes after neutralization of heparin with 1.3 mg of protamine for each 100 U of heparin, samples for ACT, arterial blood gas, activated partial thromboplastin time, PT, fibrinogen, platelet count, and thromboelastogram were collected. Laboratory tests were performed using standard referenced methods.
Ten minutes after protamine administration, the surgeon characterized the operative field as dry, moderate, or wet. Without a surgical cause for excessive bleeding, blood products were administered at the discretion of the anesthesiologist and surgeon on the basis of the likelihood of microvascular bleeding, risk factors for excessive bleeding, and coagulation test results. Fibrin glue was frequently applied to the surface of the heart.
Intraoperative autologous blood and PRBC were transfused to maintain a hematocrit greater than 30% after separation from CPB. Blood from the surgical field was collected in a dedicated canister and subsequently processed to a hematocrit of 55% with a standardized pediatric bowl (65 mL/bowl) using a Medtronic AT-1000 (Medtronic Inc, Parker, CO) to generate intraoperative autologous blood and estimate intraoperative blood loss.
Blood loss in the intensive care unit (ICU) was recorded as mediastinal chest tube drainage (MCTD) and monitored during the initial 24 hours. Mediastinal chest tube drainage was not reinfused. The cardiac surgical service directed all transfusions postoperatively on the basis of the rate of MCTD and results of the following laboratory tests: hemoglobin, PT, activated partial thromboplastin time, and platelet count.
Variables recorded include patients age; sex; height; weight; preoperative medications; heart defect; operative procedure; anesthetic medications; total heparin and protamine given; duration of CPB, aortic cross-clamp, and circulatory arrest; total fluids administered; minimal intraoperative temperature; assessment of surgical field hemostasis; transfusion requirements for the intraoperative and initial 24-hour period in the ICU; duration of intubation; 24-hour MCTD; and ICU duration.
Statistical analysis
Sample size was based on a pilot study of infants weighing less than 10 kg who underwent open heart operation and received either FFP or albumin 5% in their prime. From the findings of this pilot investigation, a sample size of 28 patients in each group would provide statistical power of 80% to detect a 30 mL/kg difference in mean 24-hour ICU blood loss between the groups.
Transfusion requirements (units) and 24-hour ICU MCTD (milliliters per kilogram) were compared between groups using the rank sum test. Because of possible differences related to complex versus simple open heart operations and to the presence of cyanosis, post hoc analyses were performed to assess differences in total transfusion requirements and ICU blood loss between infants and children who received FFP or albumin 5% in the prime after adjusting for these factors. In all cases, two-tailed p values of 0.05 or less were considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
The hemoglobin concentration on initiation of CPB, at the end of CPB, at the point of surgical hemostasis, and on arrival in the ICU was not different between the two groups.
There were 2 patients (1 albumin, 1 FFP) who required subsequent surgery but no patient in either group required mediastinal exploration for excessive bleeding. There was no difference in the time to extubation and ICU duration between the two groups. Perioperative mortality was 3.6% in each group.
| Comment |
|---|
|
|
|---|
Cyanotic heart disease [3, 17] and complex operations [18] are factors historically associated with excessive bleeding, so post hoc multivariate analyses were performed to identify any differences in blood loss between patients with either characteristic. Patients who received FFP instead of albumin 5% in the prime, either cyanotic or undergoing complex operations, experienced significantly less MCTD but not transfusion requirements. In accord with earlier studies [3, 17], cyanotic patients in this study, irrespective of prime type, experienced significantly more blood loss and greater transfusion requirements compared with acyanotic patients. Because excessive blood loss in cyanotic children is believed to occur secondary to reduced platelet count, impaired platelet aggregation, and key clotting factor deficiencies [7], FFP in the prime of cyanotic patients would merit consideration. Although FFP in the prime reduced blood loss in patients with cyanosis or complex surgery, it was associated with greater transfusion requirements as well. It is not uncommon for infants undergoing open heart operation to have reduced bleeding without improved transfusion requirements partially because of a greater likelihood of empirical transfusion than older patients [19]. However, greater donor exposure with FFP compared with albumin 5% weakens the argument for FFP in the prime.
Severe hemodilution on initiation of CPB in patients profoundly affected laboratory coagulation tests in the present study similarly to Kern and coworkers [3] in that clotting factor concentrations were considerably reduced. Patients who received albumin 5% instead of FFP in the prime in our study had significantly more prolonged PTs and lower fibrinogen concentrations immediately after administration of protamine; however, differences in these coagulation tests between the groups were absent on admission to the ICU. Neither choice of priming fluids appeared superior with respect to platelet count during CPB. Immediately after separation from CPB, the mean platelet count for patients receiving albumin 5% or FFP in the prime was similar (113.1 ± 62.3 x 109/L and 99.5 ± 50.0 x 109/L, respectively) but noticeably lower than patients who received fresh whole blood (171.2 ± 8.5 x 109/L) in Manno and associates [8] The significance of platelet abnormalities compared with clotting factor deficiencies is unclear for infants subjected to severe hemodilution. Guay and Rivard [7] found reduced platelet aggregation in infants weighing less than 10 kg as well as low concentration of clotting factors. More recently, platelet count was identified as the most significant factor associated with intraoperative blood loss in a prospective evaluation of 494 pediatric patients who underwent open heart operation [4]. Similarly, Manno and colleagues [8] attributed reduced transfusions and blood loss in patients receiving fresh whole blood to better functioning platelets.
Administration of FFP in the CPB prime or in response to abnormal coagulation values is not well defined in children [4]. However, the risks of FFP transfusion are not negligible [20] even with inactivated product [21]. Recently, FFP has been associated with worsening excessive bleeding in infants after CPB [22]. Nothing in our study suggested a worsening of coagulation status with FFP as a component of the prime compared with albumin 5%, but those patients did receive more platelet transfusions than patients who received albumin 5%, although not statistically significant (p = 0.069). Although the difference in platelet utilization may account for the difference in bleeding between the groups instead of the prime type, platelets were administered in response to a "blinded" subjective assessment of bleeding. Cryoprecipitate has been previously suggested as a substitute for FFP in the prime to achieve a greater fibrinogen concentration in these patients with less additional volume than FFP [3]. Although patients who received FFP in the prime instead of albumin 5% had a significantly greater fibrinogen concentration; the impact on hemostasis is unclear. Conceivably, because FFP does not contain a concentrated amount of any one clotting factor, as exists in cryoprecipitate regarding fibrinogen [22], significant improvement in the PT and fibrinogen concentration do not result in discernible clinical benefit even compared with albumin 5% that is devoid of any coagulation factors. The major defect in these infants and children, platelet dysfunction, is more likely to be improved by the addition of fibrinogen in cryoprecipitate [23].
Beyond the ability of albumin 5% and FFP to expand the extracellular fluid volume and maintain plasma oncotic pressure, improvement in platelet function in association with CPB has been postulated [24]. Although albumin 5% contains no clotting factors, it is believed to improve platelet function by coating the oxygenator of the CPB circuit with a protein layer that delays fibrinogen adsorption, resulting in reduced surface activation of platelets [25]. More recently, Boks and colleagues [26] demonstrated no benefit concerning platelet function after the addition of albumin in the prime, as evidenced by a lack of ß-thromboglobulin detection between albumin and control groups. In adults, FFP has been shown not to improve platelet function after CPB [27]. A similar finding was reported with FFP by Trimble and associates [28] in both adults and pediatric patients who prophylactically received two or one units of FFP, respectively. However, a beneficial effect with FFP may be more plausible in pediatric patients because of the greater hemodilution and resultant clotting factor deficiencies compared with adults undergoing CPB.
This study is limited by the lack of a transfusion algorithm for blood products; however, algorithm-based transfusion has not been developed as extensively for children as for adults [29]. Another concern involves the lack of statistical power to make a definitive conclusion regarding transfusion requirements and whether infants with cyanosis or undergoing complex operations might benefit from addition of FFP in the prime with reduced transfusion requirements as well as reduced MCTD. Finally, the degree of clotting factor replacement for each infant who received FFP is variable, possibly contributing to the lack of an identifiable hemostatic benefit. However, restricting the infants weight for inclusion in the study minimizes clotting factor concentration variability to an extent among patients. Specific clotting factor assays may have been useful to better define the benefit of FFP.
In summary, excessive bleeding and transfusion contribute to morbidity and mortality of pediatric open heart operations, with age and weight (<8 kg) recently identified as important risk factors. Blood conservation techniques are limited for infants [5, 22], so the ease and convenience of adding FFP to the prime to reduce bleeding and transfusion requirements is appealing. However, for acyanotic infants 10 kg or less undergoing noncomplex operations, this study suggests that substitution of albumin 5% for FFP in the prime will reduce transfusion requirements without increasing perioperative blood loss. Further investigation is needed to determine whether increased blood loss results in increased transfusion requirements when albumin 5% is substituted for FFP in the prime of cyanotic infants and infants undergoing complex procedures.
| Acknowledgments |
|---|
|
|
|---|
Financial support was received from the Mayo Foundation.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. C. Oliver Anticoagulation and Coagulation Management for ECMO Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2009; 13(3): 154 - 175. [Abstract] [PDF] |
||||
![]() |
M. M. McCall, M. M. Blackwell, J. T. Smyre, J. J. Sistino, J. R. Acsell, B. H. Dorman, and S. M. Bradley Fresh frozen plasma in the pediatric pump prime: a prospective, randomized trial Ann. Thorac. Surg., March 1, 2004; 77(3): 983 - 987. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |