| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesia and Intensive Care Medicine, Helsinki University Central Hospital, Meilahti Hospital, Helsinki, Finland;
Department of Anesthesia and Intensive Care Medicine, Helsinki University Central Hospital, Jorvi Hospital, Espoo, Finland; and
Finnish Red Cross Blood Transfusion Service, Helsinki, Finland
Address correspondence and reprint requests to Anne H. Kuitunen, MD, PhD, Department of Anesthesia and Intensive Care Medicine, Meilahti Hospital, Helsinki University Central Hospital, PO Box 340 (Haartmaninkatu 4), FIN-00029 HUS, Finland. Address e-mail to anne.kuitunen{at}hus.fi
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
-angle) and the strength of the fibrin clot (maximum amplitude and shear elastic modulus) were decreased up to 2 h after CPB in both HES groups. Four hours after the operation, blood loss through the chest tubes had increased in the HES groups: HES 120, mean 804 mL (range, 330–1390 mL); HES 400, mean 1008 mL (range, 505–1955 mL); and ALB, mean 681 mL (range, 295–1500 mL) (P < 0.05 between the HES 400 and ALB groups). We conclude that HES solutions, when given in doses of 20 mL/kg in connection with the CPB prime, compromise hemostasis after cardiac surgery. This effect appears related to formation of a less stable thrombus compared with that formed in the presence of ALB. IMPLICATIONS: The influence of hydroxyethyl starch (HES) on postoperative hemostasis was investigated in cardiac surgery. The thromboelastographic values indicated that HES solutions, when given in connection with the cardiopulmonary bypass prime, compromise hemostasis after cardiac surgery. This effect seems to occur through the formation of a less stable clot.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study, we compared the effects of HES and ALB on several hemostatic variables and postoperative blood loss when these solutions were used as the main component of the CPB prime solution in patients undergoing coronary artery bypass grafting. Because it has been suggested that the effects of HES on coagulation are a function of the molecular weight (MW) (6), we used two different HES solutions: one a low-MW HES preparation (MW 120,000 daltons) and the other a high-MW solution (MW 400,000 daltons). Both HES solutions had a high degree of substitution (i.e., the ratio of hydroxyethyl groups to glucose residues was 0.7). The most deleterious effects on coagulation have been attributed to large and highly substituted HES molecules (7,8).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Before surgery, the patients were allocated prospectively and in random order to one of the following groups with respect to the CPB circuit priming solution:
In all groups, acetated Ringer’s solution was added up to a final priming volume of 2000 mL.
Patients were premedicated with lorazepam 0.06 mg/kg, given orally 2 h before the induction of anesthesia. The regular oral cardiovascular medications were given at the same time as the premedication. Anesthesia was induced with fentanyl 30 µg/kg, midazolam 0.15 mg/kg, and pancuronium 0.1 mg/kg. Anesthesia was maintained until the end of operation with continuous infusions of fentanyl and midazolam (0.15 and 0.6 µg · kg-1 · min-1, respectively). When required, isoflurane was administered during the operation.
At least one arterial anastomosis (with left or right internal mammary artery or radial artery) was constructed in all patients, and additional anastomoses were completed with saphenous vein grafts. Myocardial protection consisted of an antegrade administration of crystalloid cardioplegia. The aorta was cross-clamped during suturing of the anastomoses.
During CPB, nonpulsatile pump flow and membrane oxygenation (Compactflo®; Dideco-Shiley, Modena, Italy) were used in conjunction with tepid systemic hypothermia (nasopharyngeal temperature of 32°C–34°C). Pump flow was maintained at 2.4 L · min-1 · m-2. For CPB, the patients received anticoagulation with porcine intestinal heparin 300 IU/kg, and 5000 IU of heparin was added to the priming solution. The activated clotting time (ACT) was kept >400 s during CPB. The bolus dose of protamine for neutralization of heparin effect was 1 mg/100 IU of the initial loading dose of heparin.
After CPB, the entire content of the extracorporeal circuit was collected in nonanticoagulated blood bags and returned to the patient. After surgery, the hematocrit was maintained at approximately 30% by giving packed red blood cells, acetated Ringer’s solution, or 4% ALB solution. If postoperative bleeding through the mediastinal tubes exceeded 200 mL/h, the ACT, platelet count, activated partial thromboplastin time, and prothrombin time were determined. If the ACT was prolonged >10 s as compared with the value obtained 5 min after protamine administration, a supplemental dose of protamine (20 mg) was given. If the platelet count had decreased to <100 x 109/L, 8 U of platelets (one dose) were transfused. If the activated partial thromboplastin time or prothrombin time was prolonged >1.5 times the preoperative values, 2 U of fresh frozen plasma (one dose) were transfused. If, in the bleeding patient, none of the above-mentioned criteria was fulfilled, 1 g of tranexamic acid was given.
Blood loss through the chest tubes was measured from the time of sternal closure until 16 h after surgery, and the total hemoglobin loss was calculated from a sample of the blood loss collected during this postoperative period. The amount of blood products, ALB solution, and acetated Ringer’s solution transfused and the urine output were recorded during the same postoperative period. Blood hemoglobin concentration was determined on the first postoperative morning.
The plasma creatine kinase-myocardial band (CK-MB) fraction was determined, and new Q-wave changes on the electrocardiogram (ECG) were recorded on the first postoperative morning. The durations of the intensive care unit (ICU) and hospital stay were also recorded.
Blood samples for determination of the hemostatic variables were collected in plastic syringes through a radial artery catheter. Samples were obtained before heparinization, 0.5 h after CPB, and 2 h after CPB. Samples for coagulation factor analyses were immediately cooled on ice and centrifuged, and the plasma was stored at -70°C before being assayed.
The hemoglobin and platelet counts in the whole blood were determined by using a Cell-Dyn 610 hematology analyzer (Sequoia-Turner Corp., Mountain View, CA). Platelet adhesion was studied by measuring platelet retention on glass beads (Adeplat®; Semmelweis, Milano, Italy). Template bleeding time determination was performed on the volar surface of the forearm with a Simplate II® device (General Diagnostics, Organon Teknika, Turnout, Belgium). For the platelet aggregation studies, platelet-rich plasma was prepared by centrifuging citrated blood at 150g for 10 min, and platelet-poor plasma was prepared by centrifuging blood at 1400g for 30 min. The platelet count in the platelet-rich plasma was adjusted to 100–150 x 1009/L with autologous platelet-poor plasma. Aggregation (Payton Aggregometer; Payton Associates, Scarbrough, Ontario, Canada) was induced with adenosine diphosphate (ADP). Maximal platelet aggregation was defined as the maximum increase in light transmission after the addition of ADP (30 µmol/L). The concentration of ADP causing 50% of maximum aggregation was calculated. Plasma von Willebrand factor (vWF) antigen was assayed by immunoelectrophoresis. Plasma factor VIII procoagulant activity (FVIII:C) was assayed with the one-stage kaolin-cephalin time method by utilizing FVIII:C-deficient plasma as a substrate. The FVIII:C-deficient plasma was obtained from a patient with severe hemophilia A (FVIII:C <0.01 IU/mL).
Thromboelastographic (TEG®) tracing (Hellige Co., Freiburg, Germany) was obtained from citrated whole blood. Coagulation was initiated by adding calcium chloride to the TEG® cuvettes. TEG® tracings were analyzed by measuring reaction time (r), coagulation time (r + k), clot formation rate (-angle), maximum amplitude (MA), and shear elastic modulus [G = 5000 x MA/(100 - MA)].
The sample size was estimated by using a 2-sided level of 0.05 and a power of 0.80. When a 20% decrease in the MA of TEG® tracing (TEG®:MA, i.e., the strength of the fibrin clot) with the use of HES solutions was assumed (3), the power analysis indicated that 15 patients needed to be included in each group. Analysis of variance for repeated measurements was applied within and between the groups. When significant differences (P < 0.05) were found, the Scheffé F test was used to seek significant contrasts. Single contrasts among the groups were compared by using the Mann-Whitney U-test and, in the event that discrete data were revealed, with the 
2 t-test. Associations between blood loss and hemostatic and TEG® variables were tested with analysis of simple and multiple linear regression. For the analyses, the values from all study groups were pooled. The stepwise selection procedure was applied in multiple linear regression analysis. Statistical calculations were performed with SAS software (SAS, Cary, NC). Results are given as the arithmetic means and the range.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
|
-angle) and the strength of the fibrin clot (MA and G) were depressed 0.5 and 2 h after CPB in both HES groups as compared with the ALB group. CPB caused similar hemodilution (hemoglobin concentration in whole blood) in all groups (Table 3). The hemoglobin concentration and platelet count did not differ among the groups within the entire investigation period. The adhesion decreased 0.5 h after CPB in all groups but returned to the baseline 2 h after CPB. Platelet adhesion on glass beads did not differ among the groups at any sample time. The prolongation of the bleeding time was similar in all groups. The concentration of ADP causing 50% maximum aggregation did not differ among the groups at any sample time.
|
|
In the simple regression analysis, TEG®:MA, TEG®:G, and TEG®:-angle were associated with blood loss at 4 h (R2 = 0.18, 0.20, and 0.14; P < 0.01, P < 0.01, and P < 0.05, respectively) when the values measured at 0.5 and 2 h after CPB from all groups were pooled. The association between the changes in vWF concentration from the prebypass value and blood loss at 4 h was negative (R2 = 0.13; P < 0.05). The multiple regression analysis included 15 independent variables pooled from the 3 groups: coagulation FVIII:C, vWF, TEG® variables (-angle, MA, and G) before and CPB, and their changes from prebypass values. In the stepwise procedure, three factors (baseline G, change in vWF, and change in -angle) showed best negative linear association with the blood loss at 4 h (R2 value of the model).
The clinical outcomes are shown in Table 5. One patient in the ALB group and three patients in the HES 120 group had a new Q wave in the postoperative ECG. The CK-MB fraction was increased (more than 70 ng/mL) in 3 patients each in the ALB group and the HES 120 group and in one patient in the HES 400 group. There were no differences in the length of stay in the ICU or in the hospital among the groups. None of the patients had a stroke or died.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-angle, MA, and G in the TEG® tracing, indicating the formation of a less stable thrombus than with the use of ALB. Use of HES solutions also increased postoperative blood loss, albeit significantly only in the HES 400 group. The finding of a linear correlation between the pooled values of TEG® variables and postoperative blood loss supported the hypothesis that the increased bleeding was, at least in part, due to less stable thrombus formation. The tendency to increased bleeding after cardiac surgery continues to be a problem. The bleeding is generally related to a combination of several CPB-induced alterations in the hemostatic system. In addition to the significant hemodilution that occurs with CPB, there is excessive activation of the blood coagulation system and fibrinolysis because of interaction of blood with the nonendothelial CPB surfaces (9). This study shows that the choice of the priming solution for the CPB circuit can further affect the hemostatic state.
The priming solutions used for the CPB circuit may vary widely among institutions. Colloidal solutions, including human ALB and HES, are used in an attempt to reduce fluid retention after CPB. ALB priming may be chosen because the adverse hemostatic effects of ALB seem to be limited to hemodilution (5,10). There is also an assumption that ALB may form an almost nonthrombogenic surface layer in the CPB circuit (2).
HES solutions may also impair hemostasis by mechanisms other than hemodilution. Through their coating effect, large molecules interfere with the function of the vWF and, hence, with FVIII. They also interfere with fibrin formation and, in addition, perhaps with platelet function (3,11–15). Thus, the resulting thrombus may be less stable and thereby more susceptible to lysis (3,5,16–19).
It has been speculated that the MW of HES solutions is an important factor in determining their effects on blood coagulation. Some investigations have suggested that low-MW HES solutions (120,000–300,000 daltons) may have less of an effect on coagulation than solutions with a higher MW (350,000–450,000 daltons) (5,6,11,15,20). We observed no difference between the two HES solutions in any of the numerous hemostatic indicators measured from blood and plasma.
Although it has been reported that infusion of HES just after weaning from CPB impairs surgical hemostasis to a clinically important degree (4), definitive evidence of increased postoperative bleeding with HES as a main component of priming solution is lacking. In previous studies (2,21–23), priming with either low- or high-MW HES was not associated with increased postoperative bleeding. This may have been due to the inadequate power of the studies to evaluate the problem. Although this trial was not designed to find a significant increase in postoperative blood loss, we found that CPB priming with HES solutions was associated with increased postoperative chest tube drainage—significantly so with high-MW HES. The finding is in agreement with our previous study that the use of HES in the CPB priming was associated with a tendency to increased postoperative bleeding (3). In the present study, regression analyses showed that increased postoperative bleeding was associated with formation of a less stable thrombus. The pooled values of the TEG® variables measuring clot firmness (MA, G, and -angle) showed a linear association with blood loss at four hours after operation.
The impairment of hemostasis caused by large doses of HES solutions is usually associated with a decrease in the activity of coagulation FVIII moieties, i.e., FVIII procoagulant activity, vWF antigen, and FVIII-related ristocetin cofactor (3,6,14). In this study, the decrease of FVIII:C and vWF were not observed with the use of HES solutions, which is in disagreement with our previous similar study (3). The discrepancy in the results may be explained by different study designs, i.e., different control priming solutions (ALB versus acetated Ringer’s solution) and different time points of blood samples for laboratory tests (0.5 and 2 hours versus 1 and 3 hours after CPB). In our previous study, there also appeared to be a more pronounced hemodilution after the CPB, which probably contributed to a more marked decrease in the activity of coagulation FVIII moieties. A limitation of our study is that we did not assess the functional defect in vWF by measuring ristocetin cofactor activity. However, the platelet adhesion to the glass beads measured in this trial is an indicator of interaction between vWF and platelets. On the basis of this analysis, it can be concluded that neither of the HES solutions affected this aspect of platelet function. Although there were no differences among the study groups in vWF concentrations, the changes in vWF concentrations were linearly associated with blood loss when the values from all the groups were pooled.
By means of the TEG® analysis, we were able to demonstrate that the speed at which solid clot forms (-angle) and the strength of the fibrin clot (MA and G) were decreased by HES solutions. TEG® tracings are very sensitive to even minor changes in blood coagulation, and an abnormal TEG® profile has been associated with bleeding episodes more often than abnormalities in more conventional coagulation tests (24,25). Our observation of the defect in fibrin clot formation is in agreement with other observations that hypocoagulation associated with HES solutions is caused by less stable thrombus formation rather than by a decreased amount of coagulation factors (3,5,17–20). The association of TEG® variables with blood loss in our patients further supports the utility of TEG® as a rapid point-of-care method to evaluate whole-blood clotting in the operating room (24).
In addition to the formation of solid clot, the change of -angle in the TEG® tracing was also associated with platelet function. There are conflicting reports of effects of HES solutions as CPB primes on platelet function. Boldt et al. (2) reported that HES solutions decrease platelet aggregation induced by ADP. In contrast, it has been reported that the use of different HES solutions has no additive effect on the platelet defect caused by CPB (21). In agreement with these findings, we did not observe any changes that indicated platelet dysfunction other than the decreased -angle in the TEG® tracing.
In conclusion, HES solutions as a main component of CPB prime do not appear to cause critical changes in individual coagulation factors or in platelet count and function. However, in the presence of HES solutions, the resulting thrombus is less stable, and this may cause cardiac surgical patients to be more prone to bleeding induced by shear forces or fibrinolysis after CPB.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
