| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Thoracic, Cardiac and Vascular Surgery, University of Tübingen, Germany
b Department of Anaesthesiology and Intensive Care Medicine, University of Tübingen, Germany
c Department of Medical Biometry, University of Tübingen, Germany
Received 20 September 2007; received in revised form 2 May 2008; accepted 21 May 2008.
* Corresponding author. Address: Baker Heart Research Institute, Centre for Thrombosis & Myocardial Infarction, PO Box 6492, St Kilda Road Central, Melbourne, Victoria 8008, Australia. Tel.: +61 3 8532 1490 fax: +61 3 8532 1160. (Email: Andreas.Straub{at}baker.edu.au).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
Key Words: Cardiopulmonary bypass • Deep hypothermic circulatory arrest • Haemostasis • Thromboelastometry • Congenital cardiac surgery
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
To prevent thrombotic events, blood-clotting capability is reduced during CPB by systemic anticoagulation, usually with heparin and restored after CPB by heparin neutralisation with protamine. When a bleeding tendency is observed after CPB and heparin reversal, it is usually not clear to what extent different factors such as CPB or hypothermia-induced clotting factor depletion, platelet dysfunction and rest amounts of heparin contribute to the underlying coagulopathy. Routine coagulation assays (APTT, INR) are not suitable in this situation because they do not present results quickly enough to allow evaluation as to which factors are causing the coagulopathy. As a result, empirical blood product treatment is often instituted [3]. A better understanding of the CPB and DHCA-associated coagulopathy is therefore essential for the specific treatment of each individual patient.
This study aimed to monitor haemostasis defects in infants operated on for the repair of a congenital heart defect using CPB and DHCA. Rotational thromboelastometry (ROTEM®), standard coagulation tests and platelet flow cytometry were combined and their usefulness in evaluating coagulatory function under the conditions of this study was evaluated.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
2.2 Surgery, cardiopulmonary bypass and anticoagulation management
In all patients the repair of the congenital heart defect was performed using CPB with a median duration of 90 (range: 69–113) min and a median DHCA period of 24 (range: 14–36) min. The median minimum temperature was 17.4 (range: 15.0–18.1) °C. Patients received a median amount of 4743 (range: 3200–7160) I.U. heparin for anticoagulation during CPB. None of the patients received additional heparin-derived products or antithrombin during the pre- and perioperative management. One infant who underwent repair of a tetralogy of Fallot was treated with prostaglandin E1 (PGE1) until the beginning of surgery. Nevertheless, no obvious effect of PGE1 treatment on coagulation parameters was observed in this patient.
After rewarming, patients were weaned from CPB under application of moderate doses of catecholamines. The heparin concentration was measured with the heparin management system (Medtronic, Tolochenaz, Switzerland). The amount of protamine to be administered was calculated according to the heparin level that was measured at 10 min before termination of CPB. The median amount of administered protamine was 3600 I.U. (range: 1500–5000). The median total time of surgery was 187 (range: 146–260) min. All patients received aprotinin (Trasylol, Bayer, Leverkusen, Germany) [median amount: 345,000 (range: 67,500–600,000) K.I.U.]. The priming volume for CPB consisted of 150 ml fresh frozen plasma (FFP), 150 ml packed red blood cells, 17 ml sodium bicarbonate, 20 mg/kg cephazolin and 25,000 I.U./kg aprotinin. During the rewarming phase half a unit of packed blood cells and half a unit of FFP were given and patients were haemofiltered to decrease the total pump volume and increase the haematocrit (HCT).
2.3 Blood sampling
Blood was sampled at skin incision [(1) baseline], after heparinisation during CPB [(2) directly pre- and (3) directly post-DHCA] and 5 min after protamine administration [(4) post-CPB] from a radial artery catheter or from a central venous line that had been placed for the routine anaesthetic management. Most samples were taken from the arterial catheter unless the arterial catheter was not available for blood sampling (especially in cases of technical difficulties the central venous line was used). Before blood for study purposes was collected, a discard volume was sampled from the respective catheter to exclude that measurements in study blood samples may be influenced by either dilution with fluid or by any drugs that may have been contained in the lines before sampling. At the beginning of the sampling procedure a certain amount of fluid was aspirated via the respective sampling catheter from the patient into the syringe until blood from the patient appeared in the syringe. Three times the amount of this initial volume was aspirated into the syringe before the blood sample for study parameters was taken. No additional catheterisation procedures were required for study procedures. At each sampling point 3 ml of blood were sampled and anticoagulated either with citrate for coagulation analyses and flow cytometry or with EDTA for whole blood counts.
Regarding the plasmatic coagulation, equal clotting properties can be expected in arteries and veins. The effect of different sampling sites on platelet markers has been investigated in a study involving eight patients [11]. When the data obtained from these patients were averaged only a 1.08-fold statistically not significant increase for P-selectin expression was observed in samples from the arterial line compared to the central venous line. The phenomenon of higher platelet activation marker expression in arterial blood samples can be clearer when an intraindividual analysis of single patients is performed [11] but it is obvious that regarding P-selectin expression as investigated in our study only minimal differences between arterial and central venous lines occur.
Taking these facts into account we exclude a significant effect of the different sampling sites on our coagulation and platelet data.
2.4 Standard laboratory tests
At each sampling point the activated partial thromboplastin time (APTT), the international normalised ratio (INR), fibrinogen and antithrombin III (AT III) levels as well as complete blood counts were determined in the clinical laboratory facilities of the University Hospital of Tübingen.
2.5 ROTEM® analysis
Coagulation was investigated at each sampling point using the ROTEM® thromboelastometry analyser (Pentapharm GmbH, München, Germany) according to the following principles [12]: the original thromboelastographic measurement has been modified in the ROTEM® and allows a computerised analysis of the trace. The ROTEM® contains four independent measurement channels, each providing a thromboelastometry (TEM) trace. For each ROTEM® trace, 300 µl of citrated blood are required.
By adding different coagulation-stimulating agents and/or platelet-inhibiting substances specific haemostasis defects such as coagulation factor deficiency, hypofibrinogenaemia, platelet dysfunction as well as the presence of heparin can be detected. In each ROTEM® test the clotting time (CT) and the analysis of clot formation over time indicate the dynamics of clot formation. The clot amplitude gives information about clot strength and stability, which is largely dependent on fibrinogen and platelets.
In-TEM® is a baseline test that uses an ellagic acid contact activator for analysing the general coagulatory status of the patient. With a simultaneously performed hep-TEM® test containing a heparinase, a heparin or residual heparin effect can be demonstrated when compared with the in-TEM® test. Therefore, the hep-TEM® test allows evaluating coagulatory function without a possible effect of heparin if present in the sample. The in-TEM® and hep-TEM® test, in combination, give information about whether an observed coagulopathy may be treated by application of coagulation factors or with protamine.
In the ex-TEM® test tissue factor is used as activator. Ex-TEM® is sensitive for disturbances of the extrinsic pathway of blood coagulation, for fibrinogen and fibrin polymerisation disorders, and for defective platelet function.
A differentiation between platelet function and the fibrin polymerisation process is possible with the fib-TEM® test. In fib-TEM® platelet function is eliminated with the platelet inhibitor cytochalasin D. While clots obtained in ex-TEM® or in-TEM® are composed of platelets and fibrin, the clot obtained in the fib-TEM® assay is primarily a fibrin clot. The difference between clot firmness obtained with ex-TEM® and fib-TEM® represents the platelet contribution in the ex-TEM® assay. If the fib-TEM® test reveals normal values indicating an intact fibrin network and at the same time the clot amplitude in the ex-TEM® is decreased these findings demonstrate a platelet dysfunction and possible requirement for platelet transfusion. The CT is not routinely used for diagnosis in the fib-TEM® test [13]. Therefore the clot firmness parameter A10 of the fib-TEM® assay was analysed and compared with the A10 of the ex-TEM® assay in this study.
After defined time points the amplitude of the ROTEM® trace as a measure for the clot formation is indicated, for instance, as ‘A10’ (= clot amplitude after 10 min). In order to obtain quick information about the coagulatory status of our patients we used the A10 as a measure for clot firmness in all ROTEM® assays in this study. By this approach, detailed information about a possible coagulopathy can be obtained in less then 15 min after blood sampling, which allows a quick reaction regarding blood product substitution if necessary. ROTEM® traces of one patient that are representative for measurements performed in this study are given in Fig. 1 .
|
2.7 Statistics
Analyses were performed using JMP statistical software. To compare the measurements at different time points during surgery, we performed a multifactorial analysis of variance (ANOVA) taking into account repeated measurements of a patient. The fixed factors were the different conditions and the random factor was the patient. The estimates were obtained by restricted maximum likelihood (REML). For the post-hoc comparisons among conditions, we used the Tukey-HSD test. Data were transformed into logarithms if required to stabilise variances and backtransformed after the fitting of the model. The backtransformed values are geometric means. For the overall significance level we adopted the value of 5%.
Data in the results section and in Figs. 2–4 are depicted as geometric means together with their 95% confidence intervals (95% CI) if not otherwise specified. Apart from one study by Haizinger et al. reporting CT values of infants with congenital heart disease (CHD) [14] reference values of ROTEM® data are only scarcely available for infants and children. To link our results with ROTEM® reference values as reported in the literature our data are given together with reference ranges of an adult population for the CT and A10 of in-TEM®, ex-TEM® and fib-TEM® [13] and with the CT values of in-TEM® and hep-TEM® of the report on CHD patients [14]. The reference ranges for the adult population are given as 2.5–97.5% percentiles [13] and the CT of CHD patients as means and standard deviation (SD) [14].
|
|
|
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
3.1 Clotting time of in-TEM® and hep-TEM® (Fig. 2)
3.1.1 In-TEM®
At baseline the geometric mean CT of 246 (95% CI: 194–312) s was close to values from another study on CHD patients [271 (SD: 162)] [14]. During CPB the CT was not measurable because of heparinisation. After protamine administration, the CT was distinctly elevated above the normal range (prolonged 1.8-fold in comparison to baseline).
3.1.2 Hep-TEM®
At baseline, the geometric mean CT of 218 (95% CI: 167–285) s was very close to the reference range from another study on CHD patients [mean: 219 (SD: 104)] [14]. Hep-TEM® CT was prolonged 2.1-fold directly before DHCA and 1.7-fold after DHCA. After protamine administration the CT was distinctly prolonged (1.9-fold).
The geometric mean values of the CT in in-TEM® and hep-TEM® after protamine administration lie within a similar range with a slightly higher in-TEM® CT [in-TEM®: 449 (95% CI: 355–569) s; hep-TEM®: 414 (95% CI: 317–541) s].
3.2 Amplitude of the ROTEM® trace after 10 min (A10)
3.2.1 In-TEM® (Fig. 3a)
At baseline the geometric mean A10 was 53.3 (95% CI: 46.2–61.3) mm, which is within the adult reference range (44–68 mm) [13]. During CPB the A10 was not measurable because of heparinisation. After protamine administration, the A10 was decreased 1.7-fold compared to baseline and was distinctly below the normal range.
3.2.2 Hep-TEM® (Fig. 3a)
At baseline the geometric mean A10 at 52.4 (95% CI: 48.7–56.5) mm was similar to the in-TEM® A10. The A10 was decreased 1.6-fold directly before DHCA and 1.6-fold after DHCA. After protamine administration the A10 was decreased 1.5-fold. The geometric mean values of the A10 in in-TEM® and hep-TEM® after protamine administration lie within a similar range [in-TEM®: 32.2 (95% CI: 27.9–37.1) mm; hep-TEM®: 34.5 (95% CI: 32.1–37.2) mm].
3.2.3 Ex-TEM® (Fig. 3b)
At baseline the geometric mean A10 was 52.5 (95% CI: 48.4–56.6) mm, which is within the adult reference range (43–65 mm) [13]. The A10 was decreased 1.5-fold directly before DHCA and 1.7-fold after DHCA. After protamine administration the A10 was decreased 1.4-fold and was below the normal range.
3.2.4 Fib-TEM®
At baseline the geometric mean A10 at 9.5 (95% CI: 7.8–11.6) mm was close to the lower limit of the adult reference range (9–24 mm) [13]. During CPB fib-TEM® A10 values were distinctly decreased and in some samples not measurable at all. After protamine administration the A10 was decreased 1.4-fold to geometric mean 6.8 (95% CI: 5.5–8.3) mm, which is below the reference range.
3.3 Standard coagulation measurements (APTT/INR/AT III)
At baseline the geometric mean APTT was 38.9 (95% CI: 35.2–43.0) s, which is within the normal range for infants. During CPB and heparinisation the APTT was not measurable. After protamine administration the geometric mean APTT was 81.5 (95% CI: 73.7–90.2) s, showing a 2.1-fold increase compared to baseline.
At baseline the median INR was 1.1 (range: 1–1.4). The INR was increased 3.4-fold to median 3.7 (range: 1.4–7.7) before DHCA and 2.7-fold to median 3.0 (range: 1.3–6.5) after DHCA. After protamine administration the INR was increased 1.3-fold to median 1.4 (range: 1.2–2.4) compared to baseline.
At baseline the arithmetic mean AT III value was 70.8% (95% CI: 60.8–80.6%). AT III values decreased to 56.1% (95% CI: 46.1–66.2%) before DHCA and recovered to 57.9% (95% CI: 47.9–68.0%) after DHCA and 62.9% (95% CI: 52.9–73.0%) after protamine administration.
3.4 Platelet counts
At baseline arithmetic mean platelet counts were 275 (95% CI: 242–308) x 103/µl. Platelet counts were decreased to arithmetic mean 96.6 (95% CI: 63.7–129.5) x 103/µl before DHCA, to arithmetic mean 80.2 (95% CI: 47.3–113) x 103/µl after DHCA, and to arithmetic mean 89 (95% CI: 56.1–122) x 103/µl after protamine administration.
3.5 Platelet P-selectin expression and aggregation (Fig. 4)
At baseline there was geometric mean 2.1% (95% CI: 1.4–3.2%) of P-selectin expressing platelets. During CPB P-selectin expression increased. The highest amount was observed directly after DHCA with a 2.6-fold increase to geometric mean 5.4% (95% CI: 3.6–8.2%) of P-selectin expressing platelets. With respect to a possible platelet hyporeactivity in young infants [8] platelet P-selectin expression of each patient was compared at the different sampling time points. Directly before DHCA the highest amount of P-selectin expressing platelets (25.2%) was found in the oldest infant and directly after DHCA the highest amount of P-selectin expression (31.5%) was found in the youngest infant, whereas the other patients showed an age-independent distribution regarding P-selectin expression without a tendency for an increase with older age.
Regarding platelet aggregate formation there was a slight tendency for an increase of aggregates during CPB, however without statistical significance. There was no trend of the values for each child with respect to age (data not shown).
3.6 Haematocrit
Haematocrit decreased during CPB from arithmetic mean 32.3% (95% CI: 29.0–35.6%) to arithmetic mean 28.2% (95% CI: 24.9–31.5%) pre-DHCA and arithmetic mean 27.2% (95% CI: 23.9–30.5%) post-DHCA. During the rewarming phase the haematocrit recovered to an arithmetic mean of 32.3% (95% CI: 29.0–35.6%) post-protamine.
3.7 Fibrinogen levels
Fibrinogen decreased during CPB from arithmetic mean 208 (95% CI: 173–243) mg/dl to arithmetic mean 146 (95% CI: 110–181) mg/dl pre-DHCA and 154 (95% CI: 119–189) mg/dl post-DHCA. During the rewarming phase the fibrinogen concentration recovered to an arithmetic mean of 176 (95% CI: 141–211) mg/dl post-protamine.
3.8 Blood product administration and chest drainage output
After protamine administration, a slight to moderate diffuse bleeding tendency was observed in each patient. To treat the underlying coagulopathy, which the ROTEM® results indicated to be both a defect of the plasmatic coagulation as well as platelet dysfunction, administration of fresh frozen plasma [median 55 (range 0–160) ml] as well as platelet concentrates [60 (20–160) ml] was performed, and one patient received 240 I.U. prothrombin complex concentrate (PPSB). After administration of blood products, haemostasis was satisfactory and chest closure was performed. No reoperations were required. Chest drainage output 24 h after the end of surgery was median 97 (range: 38–188) ml.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
Our data show that ROTEM® can reveal the cause of a coagulopathy within 15 min and can thereby support a quick decision on whether blood product administration with plasma or coagulation factor preparations, platelet transfusion or protamine administration should be performed.
4.1 Defect of the plasmatic coagulation
Because the hep-TEM® assay is not affected by heparinisation it can be used during CPB to predict the post-CPB coagulatory state of the patient with the advantage that specific therapy with coagulation factors and platelet transfusion can be planned and initiated in time. The fact that the CT in the hep-TEM® assay was prolonged during and after CPB in our patients indicates a generalised CPB-associated dysfunction of the plasmatic coagulation. This finding was most likely caused by a CPB-induced activation of coagulation factors and by CPB-associated haemodilution.
Because similar hep-TEM® values were observed pre- and post-DHCA, the effect of DHCA on the plasmatic coagulation adding to CPB-induced coagulopathy seems marginal in the investigated age group.
4.2 Platelet damage
During and after CPB the A10 was considerably decreased in the ex-TEM® and fib-TEM® assay. Besides the fact that the fib-TEM® decrease reflects the decrease of the fibrinogen concentration, the A10 decrease in the ex-TEM® reached its minimum directly post-DHCA, thereby suggesting a continuous loss of platelet function. This event was paralleled by an increase of platelet P-selectin that reached its maximum directly post-DHCA. This P-selectin maximum directly after DHCA may have been caused either by hypothermia and/or by contact activation of platelets in the CPB circuit preceding the hypothermic arrest period. However, the finding that the highest amount of P-selectin expression, the lowest ex-TEM® value and the lowest platelet counts are found in combination directly after DHCA suggests that hypothermia at least has an important role for platelet activation and dysfunction in our patients.
That considerable increases of platelet aggregates were not observed in our study may be explained by the fact that chilling has been reported to cluster von Willebrand (vWf) receptors on platelets, eliciting recognition of mouse and human platelets by hepatic macrophage complement type 3 (CR3) receptors. CR3-expressing but not CR3-deficient mice exposed to cold rapidly decrease platelet counts [17]. Therefore, if hypothermia had induced platelet aggregates in our patients these aggregates may have been quickly absorbed in the liver. The possibility of hypothermia-induced platelet aggregation in our patients is also supported by the finding that platelet counts were minimal directly post-DHCA, which suggests DHCA-associated activation, aggregation and consequent consumption of platelets.
As there was no direct correlation between age and P-selectin expression or aggregate formation, an obvious platelet hyporeactivity, as described for young infants [8] was not observed in our patients. Nevertheless, it should be taken into consideration that in our study platelet hyporeactivity contributed to the finding of missing CPB- and hypothermia-associated platelet aggregation.
4.3 Heparin effect
During CPB, the effect of heparin was visible in the lack of blood clotting as measured in the in-TEM® assay and in APTT measurements. After CPB and protamine administration nearly identical A10 and CT values in the in-TEM® and hep-TEM® assays revealed that heparin was successfully antagonised, excluding excessive heparin as cause for the observed coagulopathy.
4.4 Blood product application
Because ROTEM® results indicated a defect of the plasmatic coagulation as well as the platelet function after CPB, coagulation factor preparations as well as platelets were transfused to treat the observed bleeding tendency. This approach considerably improved blood clotting. At the present point of time, ROTEM® results cannot give exact recommendations on the amount of blood products that should be administered. However, ROTEM® elucidates which type of treatment may be helpful to treat a coagulopathy during CPB thereby increasing patient safety. For example it was shown in a recent report that in trauma patients ROTEM® can rapidly detect coagulation changes and might be helpful to guide transfusion [18]. Furthermore, a retrospective analysis of 990 patients showed that the use of ROTEM® statistically significantly decreases the use of red blood cells and blood products after cardiac surgery [12]. Another study reported that cumulative costs for treatment of perioperative coagulation disorders could be reduced by ‘bedside’ ROTEM® analysis to achieve a selective substitution management. This indicates that adequate differential coagulation management can also be cost-effective [19].
A comparison of the values obtained from ROTEM® baseline measurements in this study confirms that our results of the CT in in-TEM® and hep-TEM® are comparable with previous reports on infants with CHD. Furthermore, our results are close to values obtained from an adult population. These findings suggest that our results may be representative for the described patient group. However, further studies with a larger patient number are necessary to establish reference values for the parameters described in our study. In the future, the exact impact of ROTEM® on bleeding and blood product use in the setting of cardiac surgery should also be investigated, possibly by comparing patient groups that are managed with and without the use of the ROTEM®.
In our institution the ROTEM® is employed routinely for all surgical procedures involving DHCA in infants and also in adult patients. We apply this approach in order to detect a possible DHCA-related coagulopathy in its early stage so that targeted treatment with blood products can be initiated immediately. In patients that are operated on using normothermic or mild hypothermic cardiopulmonary bypass the ROTEM® is not used as a routine method unless profound coagulopathy of unclear nature persists after heparin reversal. According to our experience this approach is helpful in the treatment of a DHCA- and CPB-associated coagulopathy and may serve as a general recommendation for the use of the ROTEM® in these settings. Regarding the use of flow cytometry it should be considered that this technique can provide very helpful and detailed insights into cell function and surface receptor expression but requires specially trained staff, expensive equipment and is not easily accessible in most hospitals. In our institution flow cytometry of platelets is therefore not employed routinely but only as a clinical research method. The rationale for the use of flow cytometry in our present study was to achieve a better understanding of platelet dysfunction under in-vivo conditions of hypothermia and to further characterise the role of platelets in the setting of a DHCA- and CPB-associated coagulopathy in paediatric cardiac surgery.
In summary, according to our present experience the use of the ROTEM® alone is sufficient to evaluate the nature of the DHCA- and CPB-associated coagulopathy in the daily clinical practice.
|
5. Conclusions |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
