| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;83:1797-1803
© 2007 The Society of Thoracic Surgeons
a Departments of Anesthesiology, Perioperative and Pain Medicine, Children’s Hospital Boston, Harvard Medical School, Boston, Massachusetts
b Division of Hematology, Children’s Hospital Boston, Harvard Medical School, Boston, Massachusetts
Accepted for publication December 18, 2006.
* Address correspondence to Dr Odegard, Cardiac Anesthesia Service, 300 Longwood Ave, Children’s Hospital Boston, Boston, MA 02115 (Email: kirsten.odegard{at}childrens.harvard.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
Methods: Coagulation factors were assayed in 120 patients with CHD, divided into four age groups: group 1, 0 to 3 months; group 2, 3 to 12 months; group 3, 12 to 48 months; and group 4, older than 48 months. Healthy children without CHD were assayed as controls. Concentration of factors II, V, VII, VIII, IX, and X; protein C and S, plasminogen, and antithrombin III, were measured by standard assays. Normal ranges were determined by the empirical 95% confidence intervals.
Results: Significant reductions were found in mean levels of both procoagulant and anticoagulant factors in patients in groups 1, 2, and 3 compared with controls, but no differences were found in group 4. In group 1, all variables had significantly lower concentrations except fibrinogen and protein S; in group 2, all variables had significantly lower concentrations except for fibrinogen, factors VIII and IX, and plasminogen and protein S; and in group 3, all variables had significantly lower concentrations except fibrinogen, factors VIII and IX, and antithrombin III, plasminogen, and protein S.
Conclusions: Neonates and infants with two-ventricle CHD have lower levels of procoagulant and anticoagulant factors compared with aged-matched controls approaching normal levels in children aged older than 4 years. These coagulation factor abnormalities are similar to those described in patients with single-ventricle physiology.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Although patients with CHD may be at increased risk for thromboembolic events, the risk has predominantly been evaluated in patients with single-ventricle disease after a cavopulmonary connection. The incidence of thromboembolic events after the Fontan procedure has been reported as high as 20% to 33% [5–7]. We previously reported coagulation factor abnormalities of both procoagulant and anticoagulant proteins before and after the Fontan procedure [8–10] and, specifically, an increase in coagulation factor VIII after the Fontan operation that could predispose to thromboembolic events [10].
Whether similar abnormalities in procoagulant and anticoagulant factors exist in children with two-ventricle CHD is not known. The aim of this study was to evaluate coagulation factor values in children with two-ventricle CHD and compare maturational changes over time with healthy infants and children as age-matched controls.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Blood samples from all patients (8 mL) were obtained from an indwelling arterial catheter, when possible, after induction of general anesthesia but before surgical incision. Because the arterial and central venous catheters have continuous heparin-containing flush infusions, a standard volume of 10 mL of blood was initially drawn back from the catheter to limit possible heparin contamination of the samples. Measured indicators included prothrombin time (PT) and activated partial prothrombin time (APTT); the circulating anticoagulant factors protein C, protein S, plasminogen, and antithrombin III; and the procoagulant factors II, V, VII, VIII, IX, and X, and fibrinogen.
The Coulter T 660 (Beckman Coulter, Inc, Miami, FL) automated hematology analyzer was used to measure hemoglobin level, hematocrit, and platelets. The coefficient of variation (CV%) within and between assays at levels 1, 2, and 3 were, respectively, 1.8, 1.0, and 0.9 for hemoglobin; 1.8, 1.2, and 1.5 for the hematocrit; and 4.8, 2.7, and 1.8 for platelets.
The ACL 3000 plus (Automated Coagulation Laboratory, Beckman Coulter, Inc) was used to assay the individual procoagulant and anticoagulant factors as well as the PT and APTT. The ACL contains two measuring systems: (1) nephelometry, which is used to detect clot formation as the end point, and (2) photometry, which is used for reading chromogenic substrate assays. Blood for these assays was collected in citrated plasma (3.2% buffered sodium citrate) from an indwelling cannula from which 10 mL of blood had been aspirated to remove residual heparin. The blood was immediately centrifuged at 13,000 rpm for 5 minutes and the plasma layer removed. The PT, APTT, and fibrinogen levels were measured immediately using nephelometry (respective CV% was PT, 1.2, 2.3; APTT, 2.1, 4.8; and fibrinogen, 3.1, 5.5). Remaining plasma was stored at –70°C in 200 µL aliquots for batch performance of the other coagulation assays.
Protein C and S activity were both measured using functional clotting assays. Protein C activity was determined from the prolongation of the APTT using the Staclot Protein C kit (Diagnostica Stago, Asnieres-Sur-Seine, France) according to the manufacturer’s directions. In this assay, activated protein C inhibits factor V and VIII activity, thus prolonging the APTT of a system in which all the factors, with the exception of protein C, are present in excess; protein C is derived from the sample being tested (CV%: 1.8, 2.4).
Protein S activity was determined from the principle of factor Va inhibition using the Staclot Protein S kit (Diagnostica Stago). The principle of the test is based on the cofactor activity of protein S, which enhances the anticoagulant action of activated protein C. This enhancement is reflected by the prolongation of the clotting time of a system enriched with factor Va (CV%: 7.9, 3.8).
Extrinsic coagulation factors (factors II, V, VII, and X) were each determined by a modified PT assay. Intrinsic coagulation factors (factors VIII and IX) were determined by a modified APTT assay. In these assays, correction of the clotting time of plasma specifically deficient in the factor being tested is proportional to the concentration (activity %) of that factor in the patient plasma, interpolated from a calibration curve. Factor-deficient, plasma factors II, V, VII, VIII, IX, and X were obtained from Instrument Laboratory, Lexington, MA (CV% were II: 2.3, 2.7; V: 3.3, 2.3; VII: 2.5, 2.2; VIII: 4.9, 3.7; IX: 3.2, 3.1; and X: 3.0, 2.0).
Antithrombin III activity was determined using a synthetic chromogenic substrate assay based on factor Xa inactivation (antithrombin III, Instrument Laboratory; CV%: 4.3, 2.8). Plasma plasminogen was activated through reaction with an excess of streptokinase in the presence of fibrinogen. Plasminogen content was then determined from a synthetic chromogenic substrate assay according to the manufacturer’s directions (plasminogen, Instrument Laboratory; CV%: 4.5, 3.7).
Because altered hepatic dysfunction can contribute to coagulation factor abnormalities, serum alkaline phosphatase, 
-glutamyl transferase, alanine transaminase, aspartate transaminase, total bilirubin, albumin, and total protein were measured in all patients and compared with normal values for our laboratory.
Age-Matched Control Coagulation Variables
Maturation of the coagulation system in infants and children, termed developmental hemostasis, is well recognized [11, 12]. To establish reference ranges under the same conditions as for our clinical samples, informed written parental consent was obtained for 102 healthy infants and children undergoing minor day surgery to serve as age-matched controls. They were divided into the same age groups as described previously.
We encountered two problems obtaining samples for all control assays. First, we needed to use published control values to compare with group 1 (0 to 3 months) because of difficulties obtaining consent for blood draws from parents of healthy neonates. We used the reference range described by Andrews and colleagues [12] because it provided values for neonates and infants with similar age distributions to our patients. Second, because of technical problems, we were unable to obtain complete samples to measure PT and APPT in all of the healthy control subjects. We therefore chose to be consistent with the control data for APPT and PT across all groups and used the range of normals for age as described by Andrews and colleagues.
A total of 1.8 mL of blood was taken from each control patient after placement of a peripheral intravenous catheter and collected into citrated plasma (3.2% buffered sodium citrate). Blood samples were immediately centrifuged at 13,000 rpm for 5 minutes and the plasma stored at –70°C for subsequent batch analyses. Protein C and S, plasminogen, fibrinogen, antithrombin III, and factors II, V, VII, VIII, IX, and X were analyzed as already described.
Statistical Analysis
The sample size for each of the four age groups was determined after consideration for the wide confidence intervals for normal coagulation factor levels and that we chose a conservative significance level because of multiple comparisons. A total of 30 patients and controls in each of the four age groups was required to provide 90% power for detecting a significant difference of 1.2 standard deviations for each coagulation variable (effect size, 1.2). Sample size and power calculations were determined with the nQuery Advisor 6.0 software package (Statistical Solutions, Saugus, MA).
Control subjects and patients were compared for coagulation factors and proteins using the two-sample Student t test because the coagulation variables each conformed to a normal Gaussian-shaped distribution as evaluated by the Kolmogorov-Smirnov test. We invoked a conservative significance level of p < 0.001 to protect against committing type I errors (
error) caused by multiple testing, because more than 40 comparisons were performed (0.05/40 = 0.00125), and to provide very strong evidence of a difference [13]. Normal reference ranges derived from results of the control subjects for each age group were determined using the empirical 95% confidence intervals derived from the 2.5 and 97.5 percentile confidence limits [14]. Statistical analysis was performed with the SPSS 14.0 software package (SPSS Inc, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
|
Significant reductions were noted in mean levels of both procoagulant and anticoagulant factors in patient groups 1, 2, and 3 compared with controls, but not in patient group 4. In group 1 (0 to 3 months) concentrations were significantly lower for all variables except fibrinogen and protein S (Table 2). For group 2 (3 to 12 months), concentrations were significantly lower for all variables except for fibrinogen, factors VIII and IX, antithrombin III, plasminogen, and protein S (Table 3). For group 3 (12 to 48 months), concentrations were significantly lower for all variables except factors fibrinogen, VIII and IX, antithrombin III, plasminogen, and protein S (Table 4). No significant changes were noted in coagulation factors in children aged older than 48 months except for prolonged PT and APTT (Table 5).
|
|
|
|
All patients had oxygen saturation above 90%, and there were no abnormal liver function test results to explain the factor abnormalities. All patients recovered uneventfully from their procedures, and no sign of overt venous or arterial thromboembolic event developed before patients were discharged. Owing to the small sample size in each group, we could not determine whether there were any correlations between specific diagnosis and factor level abnormalities.
A central venous catheter was placed in 17 patients in group 1 (57%), 25 patients in group 2 (83%), 8 patients in group 3 (27%), and 10 patients in group 4 (33%). Following our routine practice, all our patients with a central venous catheter were started on an intravenous heparin infusion (10 U/[kg · hour]) within 6 to 12 hours after the cardiac procedure, or when the bleeding was under control. Most catheters were removed on the first or second postoperative day.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
The hemostatic system matures over the first years of life, leading to important differences in factor activity in children compared with adults. In general, the values of most procoagulant and anticoagulant factors are lowest in neonates and infants and increase toward adult values at varying rates, some not reaching adult levels until relatively late in the teenage years [12, 15]. Despite the observed differences, there is no evidence that healthy infants are at greater risk for hemorrhagic or thrombotic problems compared with adults, suggesting that the infant coagulation system is in functional balance at lower concentrations of most factors [16]. The use of age-matched controls is important because of this developmental hemostasis, and the reagents and type of analyzer used for factor assays also contribute to variation in the ranges of normal levels [11].
Newborns and infants younger than 1 year old have been reported to have a higher risk for vascular occlusion than older children, although in most cases an associated risk factor is present before a thromboembolic event, and the incidence decreases after the first year of life as the coagulation system matures [17]. The lower concentration of antithrombin III and protein C demonstrated in our group 1 patients, along with reduced fibrinolytic capacity and the presence of CHD as an independent risk factor, could suggest an increased risk for thrombosis in this patient population. This is speculative, however. Our study was not designed to specifically determine risk for thrombosis but, rather, to evaluate possible coagulation factor abnormalities in children with two-ventricle CHD.
The presence of an underlying medical condition, notably CHD, and placement of a central venous catheter have been cited as contributing factors predisposing to thromboembolism in children. The Canadian registry of venous thromboembolic complications in children showed that a central venous catheter was the single most important association for thromboembolic events in pediatric patients [1]. Whether the thrombophilia reported in patients with CHD is due to other physiologic derangements such as low cardiac output, venous congestion, or perhaps a genetic predisposition is unknown. Monagle [18] reported that almost 50% of infants younger than 6 months old and 30% of older children with thromboembolic disease have an underlying cardiac disorder. In a study of 171 consecutive children with venous thromboembolic events, Revel-Vilk and colleagues [19] reported that 34% of the patients had an underlying CHD. Nowak-Göttl and colleagues [17] recently studied the interaction of fibrinolysis and prothrombotic risk factors in neonates, infants, and children with and without thromboembolism and underlying cardiac disease. They found evidence of an impaired fibrinolytic system predisposing the development of early thromboembolism in these children during cardiac catheterization [17]. The rate of thromboembolic events after cardiac catheterization in pediatric patients has been estimated to be as high as 4% to 16% [17, 20, 21].
Previous studies citing CHD in general as a predisposing risk factor for thromboembolic events in the pediatric population do not clearly characterize the type of cardiac disease; that is, whether the patients have single-ventricle or two-ventricle CHD [1–3, 19, 22–25]. The increased incidence of thromboembolic events in patients with single-ventricle defects and physiology has been well demonstrated, particularly after the Fontan operation.
The cause of thromboembolic events after the Fontan procedure seems to be multifactorial. A hypercoagulable state secondary to low levels of the naturally occurring anticoagulants protein C, protein S, and antithrombin III has been postulated [7, 26, 27], and more recently, we have described elevation of factor VIII levels as a possible predisposing factor to thrombus formation [10]. However, we have also previously demonstrated that both procoagulant and anticoagulant factor level abnormalities occur earlier in the course of staged surgical palliation for patients with single-ventricle CHD (ie, preceding the Fontan operation and physiology) [8, 9].
Similarly, in this study excluding patients with single-ventricle CHD, we demonstrate that neonates and infants with two-ventricle CHD have both procoagulant and anticoagulation factor levels that are lower than age-matched controls, reaching the age appropriate level at about age 48 months. A functional procoagulant and anticoagulant balance may exist in neonates and infants with CHD, but the abnormal factor levels as shown in our study could predispose them to an acquired prothrombotic state when exposed to additional risk factors.
It has been speculated that enhanced anticoagulant properties of the vessel wall may be a factor contributing to the reduced incidence of thrombotic events in healthy children [1]. Injury of the vessel wall and endothelial damage might disrupt this protective mechanism, and inhibiting thrombin production could predispose to thrombus formation during procedures such as cardiac catheterization, cardiac operations, or central venous catheter insertion.
None of the patients in our series had a central venous catheter in place before the operation, but 50% of the patients had one placed after induction of anesthesia. No patients had overt clinical signs of a thrombosis, but this does not rule out possible thromboembolic events, and subclinical thrombus formation may have been missed because we did not routinely perform an ultrasound exam of the central veins or venography. As is routine in our institution, however, all patients with a central venous catheter were started on intravenous heparin after the operation to prevent clot formation (10 units/[kg · hour]), and the catheters are removed as soon as the patients are hemodynamically stable, commonly within the first 24 to 48 hours after the surgical procedure.
There are limitations to the interpretation of our data. In this descriptive study, we have shown that patients with two-ventricle CHD demonstrate coagulation factor abnormalities similar to those we have previously reported in patients with single-ventricle physiology before the Fontan procedure. Because of the small sample sizes in each group, however, we could not determine whether a specific cardiac diagnosis correlated with coagulation factor abnormalities.
Although our study was not designed to determine whether cyanosis or heart failure contributed to altered factor levels, all patients had an oxygen saturation level of at least 90% at the time of their operation, and none were receiving treatment for heart failure or low cardiac output state. It remains unknown whether the abnormalities we observed over time in our patients are a part of a genetic predisposition in patients with CHD in general, or result from hemodynamic or pathophysiology abnormalities in CHD patients.
Our intent in this study was to measure changes in procoagulant and anticoagulant levels over time, but our study does not provide a complete evaluation of the risk for thrombus formation because we did not perform specific platelet function studies or assess clot formation with techniques such as thromboelastography. We did not systematically monitor for thrombosis after the operation, and did not evaluate other postoperative variables such as blood product transfusion and use of antifibrinolytic drugs. It would therefore be speculative to emphasize the clinical implications of our results; a larger prospective study would be necessary to determine whether the coagulation factor abnormalities we measured support changes to clinical practice such as specific blood product replacement after cardiac operations or prolonged anticoagulation.
Finally, all patients were anesthetized at the time of blood sampling. We did not control for specific anesthetic drugs and are unable to exclude the possible effects of anesthesia on factor levels.
In summary, lower levels of procoagulant and anticoagulant factors compared with age-matched controls are demonstrated in neonates and infants with two-ventricle cardiac defects, reaching age appropriate levels after 48 months of age. The cause for these lower levels is unknown. Further investigation is needed to determine whether this predisposes to thromboembolic events in this patient population when combined with prothrombotic risk factors such as low cardiac output state, dehydration, and endothelial damage during surgical procedures, catheterization procedures, and central venous catheter insertion.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  K. C. Odegard, D. Zurakowski, J. A. DiNardo, R. A. Castro, F. X. McGowan Jr., E. J. Neufeld, and P. C. Laussen Prospective longitudinal study of coagulation profiles in children with hypoplastic left heart syndrome from stage I through Fontan completion J. Thorac. Cardiovasc. Surg., April 1, 2009; 137(4): 934 - 941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Guay Invited commentary Ann. Thorac. Surg., May 1, 2007; 83(5): 1803 - 1804. [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 |
