|
|
||||||||
Eur J Cardiothorac Surg 2007;31:866-872. doi:10.1016/j.ejcts.2007.01.033
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Department of Pediatric Cardiac Surgery, PolishAmerican Children's Hospital Collegium Medicum Jagiellonian University, 30-663 Kraków, ul.Wielicka 265, Poland
Received 2 September 2006; received in revised form 27 December 2006; accepted 22 January 2007.
* Corresponding author. Tel.: +48 12 658 10 23; fax: +48 12 657 39 47. (Email: mimalec{at}cyf-kr.edu.pl).
| Abstract |
|---|
|
|
|---|
Key Words: Fontan operation Coagulation abnormalities Liver dysfunction Thromboembolic complications
| 1. Introduction |
|---|
|
|
|---|
There has been many reports demonstrating both pro- and anticoagulant abnormalities in the late postoperative period in patients undergoing Fontan operation, but no study evaluating the status of coagulation profile and liver function in patients during staged Fontan approach in the early postoperative period [79]. The study was designed to assess the coagulation profile in children with functionally single ventricle in the early postoperative period after hemi-Fontan and Fontan procedures with the relationship to liver function and the influence of the hemodynamics on the coagulation status of blood.
| 2. Materials and methods |
|---|
|
|
|---|
All the data were derived from medical records, cardiac catheterization reports, and postoperative echocardiographic assessment. Most of the patients (97.5%) underwent previous palliative procedure suitable for the child's physiology (Norwood procedure, systemic-to-pulmonary shunt, pulmonary artery banding).
Group 1 consisted of 43 children who were qualified to the hemi-Fontan operation. The mean age of patients at the operation was 5.9 ± 3.1 months and the mean weight 6.0 ± 1.5 kg. In group 2 consisting of 37 patients, the Fontan operation was performed in the mean age of 23.2 ± 8.7 months and the mean weight of 12.5 ± 2.4 kg. Children from group 2 had previously undergone the hemi-Fontan operation in the mean age of 6.1 months. The anatomy of the congenital heart disease of children is presented in Table 1 . The hemi-Fontan operation was based on the technique described by Douville et al. [10]. The children from group 2 underwent the Fontan operation by the mean of lateral intra-atrial tunnel with the fenestration. After the hemi-Fontan and Fontan procedures all children received acetylsalicylic acid in a dose of 25 mg/kg per day from day 1 after the surgery. No patient received oral anticoagulation therapy during the time of the study. No thromboembolic event occurred in any of the patients from group 1 and 2 during the time of the study. The pleural effusions requiring drainage occurred in 7% of patients from group 1 and 46% of children from group 2.
|
|
|
|
2.1 The statistical analysis
Standard statistical methods were used. The data are expressed as mean ± standard deviation (SD). The differences between the groups concerning quantitative data were assessed using MannWhitney's test or Student's t-test as appropriate. Repeated measures ANOVA was used for the assessment of differences in coagulation factors concentration and liver function parameters. Chi-square test and its variants were used to determine the differences between the groups concerning categorical data. The relationship between coagulation factors activity and preoperative cardiac catheterization data (Table 3), postoperative variables (Table 4), liver function tests (Table 5
), and postoperative echocardiographic measurements (Table 2) was determined using multiple regression. At a p-value of <0.05 differences were considered as significant. All statistical analyses were performed using Statistica software (StatSoft, Inc., 2003, STATISTICA data analysis software system, version 6).
|
| 3. Results |
|---|
|
|
|---|
When liver function data were compared between the two groups, patients from group 1 had significantly higher aspartate aminotransferase (p < 0.0001) and patients from group 2 had significantly higher bilirubin concentration (p = 0.03) on postoperative day 1. These differences were not observed on the fifth day after the surgery. Patients from group 2 also had significantly lower total serum protein concentration throughout the postoperative course (postoperative day 1, p = 0.002; postoperative day 5, p = 0.04).
3.2 Coagulation studies
The coagulation factors abnormalities were analyzed with respect to hemostatic pediatric reference intervals [11]. The results are presented in Table 6
and Table 7
, and summarized on Fig. 1
.
|
|
|
In the same model of multiple regression linking preoperative catheterization data with the postoperative disturbances of coagulation profile, higher preoperative aortic oxygen saturation was a predictor of higher postoperative activity of factor VIII (p = 0.037) and higher preoperative single ventricle end-diastolic pressure was a predictor of lower postoperative concentration of factor VII (p = 0.03) in the Fontan group with no such influence in the hemi-Fontan group.
In the early postoperative period the general tendency was the significant decrease in the activity of prothrombin, factor VII, and antithrombin III on postoperative day 1 and the significant increase on postoperative day 5 in both groups when compared to the preoperative measure. When comparing the prevalence of coagulation factors deficiency between groups 1 and 2, significantly more patients from group 2 had a prothrombin (p = 0.0001), factor VII (p = 0.008), antithrombin III (p = 0.0048), and fibrinogen (p = 0.0025) deficiency on the first and fifth postoperative day.
In the multiple regression model taking into account postoperative hemodynamic data (heart rate, pO2, oxygen saturation, systemic venous pressure, single ventricular function, atrioventricular valve regurgitation) the concentrations of antithrombin III (p = 0.006) and factor VII (p = 0.01) were inversely related to systemic venous pressure, and prothrombin concentration (p = 0.024) was directly related to systemic pO2 in post-Fontan patients. None of the variable influenced postoperative coagulation profile in the hemi-Fontan group in this model.
As opposed to above-mentioned hemostatic parameters the concentration of factor VIII was significantly higher on postoperative day 1 in both groups compared to preoperative value (group 1, p < 0.0001; group 2, p = 0.002). On the fifth day after the surgery the concentration of factor VIII decreased significantly in patients from group 1 (p = 0.005), whereas in patients from group 2 it continued to increase. In the multiple regression there was an inverse proportion between the function of the single ventricle and the activity of factor VIII (p = 0.034) in second group with no such influence in first group.
The concentration of fibrinogen in group 1 was significantly higher on postoperative day 1 (p < 0.0001) and 5 (p < 0.0001) when compared to preoperative measure. In patients from group 2 the concentration of fibrinogen decreased significantly 24 h after the surgery (p = 0.01) and remained unchanged on postoperative day 5. In the multiply regression the fibrinogen concentration was directly related to total serum protein concentration in both hemi-Fontan (p = 0.003) and Fontan (p = 0.0002) groups.
| 4. Discussion |
|---|
|
|
|---|
The observed abnormality in liver function was an increase in aspartate aminotransferase on first postoperative day in both groups with significant dominance in the hemi-Fontan group. No serious hepatocellular damage occurred because the concentration of alanine aminotransferase was not increased significantly (Table 5) and the concentration of total serum protein and albumin was within the normal range although its significant changes throughout the perioperative period were noted. The transient increase of aspartate aminotransferase concentration observed in children after the cardiac operations in our study can be caused by cardiopulmonary bypass, myocardial damage, or stress response [12].
The bilirubin concentration was slightly increased in post-Fontan patients on postoperative day 1. This is a similar finding to other studies in which signs of mild cholestasis without serious liver dysfunction were observed in patients after Fontan operation [8,13]. The increased bilirubin level in the early postoperative period may be as well as a result of hemolysis, hypothermia, or anesthetic agents [12]. The liver function tests within normal range on postoperative day 5 in both groups without any distinction between them indicate that overall synthesizing function of the liver is preserved in spite of transient hepatic damage shortly after the surgery which may affect the coagulation profile.
The changes in total serum protein concentration and directly related to it alternations of fibrinogen content during the early postoperative period were noted in both groups. The confusing propensity of the fibrinogen to increase in the hemi-Fontan group opposite to total serum protein concentration shortly after surgery can reflect the synthesis of this protein in the liver due to acute phase reaction [21] and confirm preserved function of the liver. It was not a case in the Fontan group in which more patients had pleural effusions and an extensive loss of proteins to pleural cavities [4] with persistently lower protein concentration on the postoperative day 5 compared to previous repeated measurements as well as to the appropriate rates in the hemi-Fontan group.
We found coagulation factor abnormalities in both groups in the preoperative period, generally characterized by deficiency in factor VII, VIII, antithrombin III, and prothrombin. Those abnormalities observed also in previous studies [14,15] may result from the delayed maturation of the hemostatic system due to hypoxia, low cardiac output, or genetic disorder in patients with congenital heart disease who undergo single-ventricle palliation, but the true impact of those abnormalities on coagulation status of children remains unclear. In multiple regression linking the coagulation profile and the data from pre-hemi Fontan cardiac catheterization, patients with higher oxygen saturation in the single ventricle and pulmonary veins had a higher concentration of factor VII before the surgery. In addition, in postoperative period the concentration of prothrombin was directly related to systemic pO2 in patients after the Fontan operation. These findings confirm the important role of systemic oxygen saturation in determination of the hemostatic system function [16].
The evaluation of the hemostatic system on the first postoperative day may be difficult because of the possible changes in the concentration of coagulation factors as a consequence of acute-phase response, hemodilution, consumption coagulopathy, and blood products transfusion. Nevertheless, when comparing the coagulation profile before and after the surgery, the overall tendency in both groups was the decrease of factor VII, prothrombin, and antithrombin III 24 h after the surgery. It can suggest the global decrease of coagulation proteins concentration directly after the surgery may lead to hemorrhagic rather than thrombophilic tendency also observed in other coagulation studies [17,18].
In most patients the concentration of factor VII, antithrombin III, and prothrombin returned to normal value on the postoperative day 5. However, there were still some abnormalities in coagulation profile that could contribute to hemostatic imbalance. The concentration of factor VII was below normal values in 30% of patients after Fontan operation on postoperative day 5. It can theoretically predispose to bleeding. However, previous studies have reported that low concentration of factor VII may in fact support tissue-factor inducing coagulation in pathologic states [19]. Also there is evidence of an increased risk of thromboembolism in patients with hereditary factor VII deficiency [20]. In our study, higher single ventricle end-diastolic pressure before the Fontan operation was a predictor of lower activity of factor VII postoperatively. Also, decreased factor VII and antithrombin III concentration was observed in patients with higher systemic venous pressure after the Fontan operation. These relationships indicate that hemodynamic factors are the main reason impairing the liver synthesizing function and coagulation proteins activity which is emphasized by the fact that they were detected only in the Fontan group.
As opposed to other hemostatic parameters the concentration of factor VIII was increased 24 h after the surgery in the hemi-Fontan and Fontan patients. Also, a higher pre-Fontan aortic oxygen saturation was a predictor of higher factor VIII concentration in the postoperative period. The immediate increase in activation of factor VIII in the early postoperative period likewise activity of fibrinogen in our patients may reflect an acute-phase reaction and cause hypercoagulable state [17,21]. Patients in the Fontan group in contrast to hemi-Fontan group still had increased factor VIII concentration on postoperative day 5. This finding is similar to earlier reports in which increased factor VIII was observed in patients after the Fontan operation [22,23]. Factor VIII was found to be an independent risk factor for thromboembolic complications in the Leiden Thrombofilia Study with a five- to sixfold increased risk when the concentration is above 150% [24]. In addition, a study of Hollestelle et al. found that mostly non-parenchymal cells (liver sinusoidal endothelium), not hepatocytes, play a major role in factor VIII synthesis [25]. In our report the concentration of factor VIII above 150% occurred in more than one-third of post-Fontan patients on the first and fifth postoperative day (Table 6) and children with decreased ventricular function had a higher concentration of factor VIII. The immediate increase in systemic venous pressure after the Fontan operation could contribute to increased factor VIII concentration which is apparent in the early postoperative period. This agrees with a study of Odegard et al. who found that systemic venous pressure was strongly related to factor VIII concentration [22].
Decreased antithrombin III and factor VII concentration with a significant increased factor VIII concentration in addition to other risk factors (increased venous pressure, change in the pattern of blood flow, temporary hepatic injury, hypoxia, and single ventricle dysfunction) in patients after Fontan operation can contribute to increased incidence of thromboembolic complications in the early postoperative period. These findings agree with previous reports that imbalance of both pro- and anticoagulant factors after Fontan operation predispose to thrombosis [79,15,22,23].
4.1 The limitations of the study
In our study we did not have the age-matched control group to compare the coagulation profile of patients with the values assessed in a control cohort, but we used hemostatic pediatric references intervals from a study of Andrew et al. [11] to evaluate the abnormalities of coagulation profile in our patients.
There is no consensus about the prophylactic antithrombotic strategy in children undergoing Fontan operation [2,3]. In the study all of the patients were taking aspirin as the antiplatelet therapy from the postoperative day 1. We do not routinely use the oral anticoagulation and we did not find any evidence of thromboembolic complications throughout the whole hospital stay in our patients. This might result from a short follow-up, small number of patients, or demonstrate that aspirin is sufficient in prophylactic antithrombotic therapy. On the other hand, based on our results, we could identify a group of children with perioperative factors associated with coagulation abnormalities and possible higher risk for thrombosis and consider the option of changing aspirin to more aggressive anticoagulation therapy. In the future studies it should be assessed weather observed coagulation abnormalities in post-Fontan patients coexist with the higher incidence of thrombosis in the late postoperative period and if adjusting the prophylactic anticoagulation strategy brings the desirable effect.
In conclusion, our study demonstrated a high incidence of both procoagulant and anticoagulant factors abnormalities in patients during staged Fontan approach in the early postoperative period. The transient hepatic damage in the early postoperative period may contribute to coagulation profile abnormalities. The decreased antithrombin III, factor VII, and increased factor VIII concentrations combined with the influence of unique hemodynamics on hemostatic parameters in patients after Fontan operation can contribute to increased risk of thromboembolic complications in the early postoperative period.
| Appendix A |
|---|
|
|
|---|
Dr R. Jonas (Washington, D.C., USA): What are the clinical implications of your work? Fenestration thrombosis early after surgery can frequently have a serious hemodynamic impact. And thrombus formation within a baffle or extracardiac conduit also can be a serious early postoperative problem. What are your recommendations regarding reversal of heparin after a Fontan operation, regarding the use of aprotinin during the Fontan operation, and what are your recommendations regarding long-term management of anticoagulation particularly with reference to aspirin and warfarin or Coumadin?
Dr Procelewska: We didnt use aprotinin in any patient after the Fontan operation. According to the prophylactic anticoagulation therapy, after stopping the bleeding shortly after the surgery in the ICU, we used the continuous infusion of heparin in all our patients, which was followed on the postoperative day 1 by a low dose aspirin of 2 to 5 mg/kg. We didnt use in any patient the oral anticoagulation therapy.
Concerning the first question, the clinical implications, I think that based on our studies we could define a subgroup of patients, in whom those hemodynamic changes, both in the preoperative period and also postoperatively, exist. In those patients we could measure the profile of coagulation abnormalities. If those abnormalities persist, we might think about changing the anticoagulation therapy into the oral anticoagulation.
Mr S. Zaefs (Princeton, N.J., USA): What's your hypothesis about the correlation between central venous pressure and coagulation? What's your hypothesis, why this is connected, the parameters of venous pressure and the concentration of coagulation factors?
And the second question, you didnt show it in your presentation, but maybe you have some pilot data. What's going on with factor XIII in this category of patients?
Dr Procelewska: We didnt measure factor XIII in our study, unfortunately.
Regarding the first question, in the hemi-Fontan group I think that single ventricle function is the only factor contributing to the changes in hepatic circulation. In post-Fontan patients, the change in the function of the ventricle is another associating factor that contributes to increase in PA pressure and increased venous pressure. So maybe the changes in the hemodynamics after Fontan procedure may influence significantly the changes in coagulation protein synthesis.
We think that increased venous pressure causing hepatic congestion causes the disturbances in coagulation protein synthesis by decreased synthesis of factors which are synthesized in the cells, and increase in factor VIII which is synthesized not in the cells of the liver but they are produced by the lining cells of the sinuses of the livers.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. J. Karsies, K. K. Nicol, M. E. Galantowicz, J. A. Stephens, and B. A. Kerlin Thrombotic Risk of Recombinant Factor Seven in Pediatric Cardiac Surgery: A Single Institution Experience. Ann. Thorac. Surg., February 1, 2010; 89(2): 570 - 576. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Camposilvan, O. Milanesi, G. Stellin, A. Pettenazzo, L. Zancan, and L. D'Antiga Liver and Cardiac Function in the Long Term After Fontan Operation Ann. Thorac. Surg., July 1, 2008; 86(1): 177 - 182. [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 |