Albumin (ALB), gelatin (GEL), and hydroxyethyl starch (HES) solutions are considered effective plasma expanders in patients undergoing cardiac surgery (1,2) although deleterious effects of some colloid solutions on hemostasis may increase the risk of postoperative bleeding (3). A coagulation disorder with increased blood loss has been demonstrated, particularly when high-molecular-weight HES with a high degree of substitution (DS) is given during cardiac surgery (4,5).
In previous studies, no adverse effects were reported after GEL infusion to cardiac surgical patients (2,4,6). After total hip or knee replacement, whole blood coagulation analysis indicated postoperative hypercoagulation (7), and the increase of factor VIII and von Willebrand factor concentration was not blunted by GEL (8). Conversely, 20 or 30 vol.% whole blood hemodilution with GEL in vitro decreased clot strength assessed by thromboelastometry, although the impairment was somewhat less than with low molecular weight HES (MW 120,000–200,000) (9,10). GEL also impaired platelet aggregation and final clot formation in healthy volunteers (11) and during cardiac surgery (12).
In clinical practice, colloids are often used immediately after cardiac surgery to maintain left ventricular preload (13) (i.e., during the period of increased risk for bleeding due to cardiopulmonary bypass [CPB] induced coagulation defects.) Because little is known about this immediate postoperative administration of colloids, we tested the hypothesis that the effect of GEL or HES is more than that of ALB on blood coagulation assessed by thromboelastometry after on-pump cardiac surgery.
Forty-five patients scheduled for elective primary and single cardiac surgery were included in the study. The Institutional Ethical Committee approved the study protocol, and all patients gave written informed consent to participate in the study. Patients with preoperative coagulation disorders, renal or hepatic failure, or taking medication with coumarin anticoagulants, heparin, and/or acetylsalicylic acid within the previous 5 days were excluded.
Patients were premedicated with lorazepam, and the regular oral cardiovascular medications were given. For induction of anesthesia, the patients received fentanyl or sufentanil and propofol or etomidate. Pancuronium or rocuronium was used as muscle relaxant. Anesthesia was maintained with a continuous infusion of propofol and fentanyl or sufentanil until the end of surgery. Isoflurane supplementation was used to achieve a bispectral index level below 60.
During CPB, the nasopharyngeal temperature was 32°C–30°C. The extracorporeal circuit priming consisted of 2000 mL of Ringer’s solution and 100 mL of 15% mannitol. For CPB, the patients received heparin 300 IU/kg, and 5000 IU of heparin was added to the priming solution. The activated clotting time (ACT) was measured every 30 minutes and kept above 480 s during CPB. An additional dose of 5000 IU of heparin was given if required. During CPB, the hematocrit was kept above 20%. The dose of protamine was 1 mg/100 IU of the initial loading dose of heparin for the neutralization of the effect of heparin after CPB. Additional doses of 25 mg of protamine were given to achieve the prebypass ACT level. After CPB, blood from extracorporeal circuit was collected into nonanticoagulated bags and transfused. Tranexamic acid, ε aminocaproic acid, aprotinin, ALB, GEL, or HES solutions were not given during operation.
Immediately after admission to cardiac surgical intensive care unit (ICU), the patients were allocated in random order (closed envelopes) to receive one of the following infusions:
- 4% ALB solution, 15 mL kg−1 (4% albumin SPR®; Finnish Red Cross Blood Transfusion Service, Helsinki, Finland) (ALB group, n = 15)
- 4% succinylated GEL solution, 15 mL kg−1 (Gelofusine®; 40 mg mL−1; B. Braun, Melsungen, Germany) (GEL group, n = 15)
- 6% HES solution, 15 mL kg–1 (HAESSteril®; 60 mg mL−1, average molecular weight 200 kDa, molar substitution ratio 0.5; Fresenius Kabi, Bad Homburg, Germany) (HES group, n = 15).
The infusion rate was clinically adjusted to optimize cardiac filling pressures and was based on serial hemodynamic measurements. Hemodynamic measurements were recorded using radial and pulmonary artery catheters. Cardiac index was measured by thermodilution technique in triplicate. Mean arterial blood pressure was maintained above 70 mm Hg, pulmonary artery occlusion pressure at 10–14 mm Hg (based on the echocardiographic measurements at the end of surgery), and cardiac index at >2.0 L min−1 m−2. Norepinephrine or epinephrine infusion was started when needed. In the ICU, hemoglobin was maintained at or more than 8.0 g dL−1 with red blood cell concentrates (RBC). After completion of the study infusion acetated Ringer’s solution was transfused when required. If postoperative blood loss exceeded 200 mL h−1, ACT, platelet count, activated partial thromboplastin time (APTT), and prothrombin time (PT) were determined. If ACT was prolonged more than 10 s compared with the prebypass value, a supplemental dose of 25 mg of protamine was given. If the platelet count was less than 100 × 109 L−1, 1 unit/10 kg platelet concentrate was administered. If the APTT or PT was prolonged more than 1.2 times compared with preoperative values, 10 mL kg−1 fresh frozen plasma (FFP) was transfused. If bleeding continued, but none of the above-mentioned criteria were met, 1 g of tranexamic acid was given.
Blood samples for thromboelastometry were collected via a nonheparinized radial artery catheter into polypropylene tubes (Vacuette®; Greiner Bio-One, Kremsmueter, Austria) containing 3.2% buffered citrate before the administration of the study colloid (Pre), immediately after completion of the study infusion (Post), and 2 hours after completion of the study infusion (2 h). Hemoglobin concentration, hematocrit value, platelet count, ACT, lactate, and calcium ion concentration were also measured Pre, Post, 2 h, and on the first postoperative morning (1.POM).
Modified thromboelastometry coagulation analysis (ROTEM®; Pentapharm CO, Munich, Germany) using four activators [intrinsic ROTEM (InTEM®) = In; extrinsic ROTEM (ExTEM®) = Ex; fibrinogen ROTEM (FibTEM®) = Fib; native ROTEM (NaTEM®) = Na] was performed simultaneously by an investigator (TN or AK) blinded to the study colloid (6,14). Tests, definitions, and normal values of variables of thromboelastometry (ROTEM®) are presented in Table 1 (15). In the FibTEM®, all platelet function has been removed by adding cytochalasin D, which destroys the cytoskeleton of the platelets in the blood sample. Thus, the formed clot measures the quality of fibrin polymerization. The coagulation was initiated with activators by using a semiautomated electronic pipette system according to the manufacturer’s instructions. Coagulation was allowed to proceed 60 min. Automatic ROTEM® variables were: coagulation time (CT, s), clot formation rate (CFR, s), α-angle (α, degree), and maximum clot firmness (MCF, mm). Furthermore, shear elastic modulus (G = 5000 × MCF [100 – MCF] – 1) was calculated. The effect of platelets on clot strength was assessed by the difference between ExTEM®- and FibTEM®-induced MCF: Platelet MCF = ExMCF – FibMCF (14).
The hemoglobin concentration, hematocrit value, and platelet count in whole blood were determined using Cell-Dyn 610 hematology analyzer (Sequila-Turner Corp., Mountain View, CA). ACT was measured by ACT II® device (Medtronics, Inc., Minneapolis, MN). Plasma levels of lactate and ionized calcium were analyzed by an ABL 700® device (Radiometer, Copenhagen, Denmark).
The cumulative chest tube drainage, urine output, and the amount of blood products and acetated Ringer’s solution transfused were recorded at the arrival to ICU, at the end of the study infusion, and 2 hours later as well as on the first POM.
The number of patients needed was based on an expected difference in MCF of the thromboelastometry tracing. Based on our previous study, 15 patients per group were considered necessary to detect statistical significance with an α- and β-error of 0.05 and 0.2, respectively (5). In that study (5), data were considered to be normally distributed, and power analysis in the present study was performed by using arithmetic means with standard deviations (SD). The differences between and within the groups were analyzed with one-way analysis of variance on ranks because Normality and Equal variance tests were not passed in every case indicating skewed data. The Student-Newman-Keuls test was used for paired comparisons. The results are reported as medians and range unless otherwise indicated. The frequencies were tested by χ2-test or Fisher’s exact test as appropriate. Correlations between blood loss at 2 h or 1.POM and ROTEM® variables were modeled with analysis of simple linear regression and multiple backward stepwise linear regression. For these analyses, the values from all the study groups were pooled. All the statistical calculations were performed using the SigmaStat® for Windows (version 2.03; SPSS Inc., Chicago, IL).
Fifty-four patients gave their informed consent to participate in the study. Nine patients were excluded because of use of antifibrinolytic drugs during surgery, off pump technique, or postponed surgery.
Patients in the ALB, GEL, and HES groups were similar regarding demographic and preoperative data (Table 2). Patients’ routine preoperative laboratory screening tests were normal (data not shown) and preoperative left ventricular ejection fraction was >40%. Surgery-related data are presented in Table 2. Five patients in the ALB group, four in the GEL group, and three in the HES group received RBC concentrates intraoperatively. FFP or platelets were not needed intraoperatively. Urine output was similar among the study groups (data not shown).
There were no significant differences in the administration times of the study infusions (Table 3). The dilutional effect on hemoglobin and platelet count is presented in Table 3. Despite the mild thrombocytopenia, the ROTEM® values indicating platelet function (platelet MCF = ExMCF- FibMCF) remained unaffected in all groups.
The coagulation time (ExCT) indicating initial fibrin formation was prolonged immediately after HES infusion but not after ALB or GEL (Table 4). Coagulation times obtained by InTEM® and NaTEM® were not different among the groups at any study point (data not shown). Fibrin buildup was most disturbed (prolonged InCFT, ExCFT, and decreased α-angle) immediately after completion of HES infusion (Table 4). At the same time, a somewhat milder but significant prolongation in InCFT and ExCFT and a decrease in α-angle were observed in the GEL group but not in the ALB group. InCFT was also longer and In α-angle more depressed at 2 h in the GEL and HES groups compared with the ALB group. Calcium induced CFT (NaCFT) and α-angle did not differ among the groups at any study point (data not shown).
MCF (all activators of ROTEM®) was decreased immediately after the completion of GEL and HES infusions but remained unchanged after ALB (Fig. 1, A–D). Two hours after completion of the infusion, InMCF was still decreased in the GEL and HES groups (Fig. 1A). FibMCF and NaMCF were more dcreased in the HES group than in the GEL group immediately after the completion of the infusion (Fig. 1, C and D). A decrease in FibMCF and NaMCF was still seen at 2 h in the HES group. In the GEL group, they were partly recovered. Changes of shear elastic modulus (G) were parallel with MCF during the investigation (Table 5).
There were no reoperations as a result of bleeding or any other reason. Median (range) cumulative chest tube drainage during the study was: ALB Group, 840 mL (510-1850); GEL Group, 1070 mL (410-2100); HES Group, 1140 mL (580-1800); P = 0.28 among all groups.
The number of transfused RBC, FFP, or platelets were not different among the study groups. When the number of all blood products were pooled, a larger proportion of patients in the HES group than in the other groups received blood products (ALB Group, 33%; GEL Group, 20%; and HES Group, 67%; P = 0.027 among all groups). The number of patients who received additional protamine or tranexamic acid postoperatively was two, three, four and one, two, three in the ALB, GEL, and HES groups, respectively.
In the simple regression analysis, ROTEM®: In α-angle at sample Pre; ROTEM®: ExCFT and ExMCF at sample Post pooled from all groups were linearly associated with chest tube drainage at 2 h (r = 0.347, 0.321, and 0.294; P = 0.02, P = 0.03, and P = 0.05, respectively). Total chest tube drainage (1.POM) associated significantly only with ROTEM®: ExMCF at sample Post (r = 0.298; P = 0.046). When analyzing each group separately by simple regression analysis ROTEM®: ExMCF pooled from GEL and HES groups showed negative linear association with chest tube drainage at 2 h (r = 0.4; P = 0.027) and with total chest tube drainage (r = 0.392; P = 0.032) (Fig. 2, A and B). In the ALB group such an association was not observed (chest tube drainage at 2 h r = 0.177 and P = 0.528; total chest tube drainage r = 0.049 and P = 0.863).
In the backward multiple stepwise regression analysis (a total of 30 possible explanatory ROTEM® variables at sample Pre and Post pooled from three groups), the total chest tube drainage could be predicted from a linear combination of the ROTEM® variables indicating fibrin formation (InCT, ExCT, ExCFT, and NaMCF; P = 0.025, P = 0.018, P = 0.003, and P = 0.022, respectively) at sample Pre and clot strength (ExMCF and ExG; P = 0.005 and P = 0.049, respectively) at sample Post.
Our results demonstrate that both GEL and HES solutions impair hemostasis when infused immediately after cardiac surgery. The effect of postoperative administration of colloids on whole blood coagulation has not been studied after cardiac surgery. In our study, succinylated GEL and HES 200/0.5, but not ALB, induced a hypocoagulable state, as indicated by prolonged CFT, decreased α-angle, MCF, and G in the thromboelastometry tracing (i.e., a less stable thrombus). The postoperative chest tube drainage was not statistically different among the three groups, but the linear correlation between the pooled values of thromboelastometry variables (at samples Pre and Post) and postoperative chest tube drainage support the idea that the increased bleeding tendency was due to a less stable thrombus formation.
Colloids are often used during and after surgery because of their ability to maintain intravascular fluid volume and regional tissue perfusion more efficiently than crystalloids (16). Extracorporeal circulation induces platelet dysfunction, reduces the amount of coagulation factors, and promotes fibrinolytic activity (17). The effects of colloid solutions administered soon after CPB on blood coagulation may therefore be of importance. The current study demonstrates that the choice of colloid solution may affect the postoperative hemostatic state.
Our results support the idea that the hemostatic effects of ALB seem to be limited to hemodilution, because thromboelastometry assessed coagulation was best preserved after ALB infusion. In addition, there were no signs of the experimentally demonstrated mild ALB-hemodilution related hypercoagulability in the present study (10,18). The statistically different hemoglobin concentration decrease between the ALB and GEL or HES groups immediately after completion of the infusions may be related to the different volume effects of the test solutions and may have affected our results (19,20). The somewhat larger chest tube drainage in both artificial colloid groups compared with the ALB group may also partly explain the difference. Hemoglobin concentration hardly affected thromboelastometry variables because the patterns of the groups were different at comparable levels of hemoglobin at 2 hours after completion of the infusion.
GEL reduced the quality of clot formation in the current study, which is in agreement with previous in vitro and in vivo findings (9–11,21). The mechanism of the GEL-induced hypocoagulable state seems to be comparable with that of HES as the patterns of thromboelastometry tracings are quite similar (21). It is possible that GEL molecules (MW 30,000) interfere with the function of coagulation factors through a coating effect, because CFT, α-angle, and MCF, by both intrinsic and extrinsic coagulation pathways (ExTEM®, InTEM®), were impaired up to 2 hours in the GEL group. Fibrinogen-dependent MCF (FibTEM®) was also clearly depressed after GEL, indicating disturbance in the quality of fibrin polymerization. The previously suggested primarily platelet-mediated mechanism of GEL-induced impairment of hemostasis (22) is not supported by our trial, because MCF, after excluding the effect of fibrinogen, remained unchanged. Our results are also in contrast to in vitro studies with no change in thromboelastometry variables after GEL hemodilution and to some clinical studies in which GEL had no effect on hemostatic variables and did not increase blood loss after surgery (8,23,24). The discrepancy between these clinical trials and the present study may be due to our relatively rapid (median, 95–105 min) administration of colloids and the fact that after CPB, hemostatic mechanisms per se are disturbed for up to 2 h postoperatively, and the coagulation process is therefore vulnerable (17).
The administration of HES solution impairs hemostasis mainly by decreasing clot strength (5,10,21). In the current study, HES solution both slowed clot formation and decreased MCF. The mechanism of impaired hemostasis by HES molecules may therefore be due to compromise of coagulation factors, such as thrombin-fibrinogen and factor XIII-fibrin polymer interactions, rather than induced platelet dysfunction (25,26). HES solution in CPB priming has been postulated to affect platelet function. Boldt et al. (27) reported HES solutions to decrease adenosine diphosphate-induced platelet aggregation, but HES had no impact on platelet retention on glass beads, bleeding time, or platelet aggregation in our previous study (5). As expected, HES 200/0.5 induced somewhat more profound alterations in thromboelastometry tracing than GEL in the present study. The findings are in accordance with trials reporting a less negative influence of GEL on hemostasis compared with some HES preparations. These in vitro investigations studying mild to moderate whole blood dilution with HES (HES 450/0.7; HES 200/0.5; HES 130/0.4, or HES 120/0.7) or GEL, have shown significantly greater impairment in Sonoclot or thromboelastometryassessed coagulation in HES-diluted samples (9,10,28). Despite the compromised coagulation after GEL- or HES-dilution in in vitro studies, in cardiac surgical patients infusion of GEL or HES 130/0.4 seems to be equally safe (6). As MW and degree of substitution of a HES preparation increases and the administration occurs intra- and postoperatively, the total blood loss may be larger compared with GEL in patients undergoing cardiac surgery (2). However, if HES 200/0.5 is not given as a rapid bolus-like infusion, the negative effects on coagulation may be less harmful (4,29).
We could not show any statistically significant difference in postoperative chest tube drainage, which is in accordance with the study of van der Linden et al. (2). We did find a significant linear correlation between clot strength impairment and chest tube drainage. This suggests that the bleeding was due to GEL or HES reduction of stable thrombus formation. We could also predict chest tube drainage using our multiple regression model: A combination of thromboelastometry variables indicating impaired fibrin formation and clot strength before and after the administration of test solutions. This confirms the clinical efficacy of thromboelastometry in assessment of coagulation disorders after cardiac surgery (30).
In summary, we conclude that immediate administration of either GEL or HES 200/0.5 induces whole blood coagulation abnormalities in fibrin formation and clot strength after cardiac surgery with CPB. This impairment in hemostasis may predispose patients to increased bleeding tendency.
We are grateful to Edward Munsterhjelm, MD, for assistance in preparing the illustrations.
1. Tølløfsrud S, Svennevig JL, Breivik H, et al. Fluid balance and pulmonary functions during and after coronary artery surgery: Ringer’s acetate compared with dextran, polygeline, or albumin. Acta Anaesthesiol Scand 1995;39:671–7.
2. Van der Linden PJ, De Hert SG, Daper A, et al. 3.5% urea-linked gelatin is as effective as 6% HES 200/0.5 for volume management in cardiac surgery patients. Can J Anaesth 2004;51:236–41.
3. Wilkes MM, Navickis RJ, Sibbald WJ. Albumin versus hydroxyethyl starch in cardiopulmonary bypass surgery: a meta-analysis of postoperative bleeding. Ann Thorac Surg 2001;72:527–33.
4. Boldt J, Knothe C, Zickmann B, et al. Influence of different intravascular volume therapies on platelet function in patients undergoing cardiopulmonary bypass. Anesth Analg 1993;76:1185–90.
5. Kuitunen AH, Hynynen MJ, Vahtera E, Salmenpera MT. Hydroxyethyl starch as a priming solution for cardiopulmonary bypass impairs hemostasis after cardiac surgery. Anesth Analg 2004;98:291–7.
6. Haisch G, Boldt J, Krebs C, et al. Influence of a new hydroxyethyl starch preparation (HES 130/0.4) on coagulation in cardiac surgical patients. J Cardiothorac Vasc Anesth 2001;15:316–21.
7. Karoutsos S, Nathan N, Lahrimi A, et al. Thrombelastogram reveals hypercoagulability after administration of gelatin solution. Br J Anaesth 1999;82:175–7.
8. Huttner I, Boldt J, Haisch G, et al. Influence of different colloids on molecular markers of haemostasis and platelet function in patients undergoing major abdominal surgery. Br J Anaesth 2000;85:417–23.
9. Egli GA, Zollinger A, Seifert B, et al. Effect of progressive haemodilution with hydroxyethyl starch, gelatin and albumin on blood coagulation. Br J Anaesth 1997;78:684–9.
10. Niemi TT, Kuitunen A. Artificial colloids impair haemostasis. An in vitro
study using thromboelastometry coagulation analysis. Acta Anaesthesiol Scand 2005;49:373–8.
11. de Jonge E, Levi M, Berends F, et al. Impaired haemostasis by intravenous administration of a gelatin-based plasma expander in human subjects. Thromb Haemost 1998;79:286–90.
12. Tabuchi N, de Haan J, Gallandat Huet RC, et al. Gelatin use impairs platelet adhesion during cardiac surgery. Thromb Haemost 1995;74:1447–51.
13. Karanko MS, Klossner JA, Laaksonen VO. Restoring volume by crystalloid versus colloid after coronary artery bypass: hemodynamics, lung water, oxygenation, and outcome. Crit Care Med 1987;15:559–66.
14. Boldt J, Haisch G, Kumle B, et al. Does coagulation differ between elderly and younger patients undergoing cardiac surgery? Intensive Care Med 2002;28:466–71.
15. Lang T, Bauters A, Braun SL, et al. Multi-centre investigation on reference ranges for ROTEM thromboelastometry. Blood Coagul Fibrinolysis 2005;16:301–10.
16. Lang K, Boldt J, Suttner S, Haisch G. Colloids versus crystalloids and tissue oxygen tension in patients undergoing major abdominal surgery. Anesth Analg 2001;93:405–9.
17. Kuitunen AH, Heikkila LJ, Salmenpera MT. Cardiopulmonary bypass with heparin-coated circuits and reduced systemic anticoagulation. Ann Thorac Surg 1997;63:438–44.
18. McCammon AT, Wright JP, Figueroa M, Nielsen VG. Hemodilution with albumin, but not Hextend, results in hypercoagulability as assessed by thrombelastography in rabbits: role of heparin-dependent serpins and factor VIII complex. Anesth Analg 2002;95:844–50.
19. Jones SB, Whitten CW, Despotis GJ, Monk TG. The influence of crystalloid and colloid replacement solutions in acute normovolemic hemodilution: a preliminary survey of hemostatic markers. Anesth Analg 2003;96:363–8.
20. Jones SB, Whitten CW, Monk TG. Influence of crystalloid and colloid replacement solutions on hemodynamic variables during acute normovolemic hemodilution. J Clin Anesth 2004;16:11–7.
21. Innerhofer P, Fries D, Margreiter J, et al. The effects of perioperatively administered colloids and crystalloids on primary platelet-mediated hemostasis and clot formation. Anesth Analg 2002;95:858–65.
22. Evans PA, Heptinstall S, Crowhurst EC, et al. Prospective double-blind randomized study of the effects of four intravenous fluids on platelet function and hemostasis in elective hip surgery. J Thromb Haemost 2003;1:2140–8.
23. Mortier E, Ongenae M, De Baerdemaeker L, et al. In vitro evaluation of the effect of profound haemodilution with hydroxyethyl starch 6%, modified fluid gelatin 4% and dextran 40 10% on coagulation profile measured by thromboelastography. Anaesthesia 1997;52:1061–4.
24. Beyer R, Harmening U, Rittmeyer O, et al. Use of modified fluid gelatin and hydroxyethyl starch for colloidal volume replacement in major orthopaedic surgery. Br J Anaesth 1997;78:44–5.
25. Fenger-Eriksen C, Anker-Møller E, Heslop et al. Thrombelastographic whole blood clot formation after ex vivo
addition of plasma substitutes: improvements of the induced coagulopathy with fibrinogen concentrate. Br J Anaesth 2005;94:324–9.
26. Nielsen VG. Colloids decrease clot propagation and strength: role of factor XIII-fibrin polymer and thrombin-fibrinogen interactions. Acta Anaesthesiol Scand 2005;49:163–71.
27. Boldt J, Zickmann B, Ballesteros BM, et al. Influence of five different priming solutions on platelet function in patients undergoing cardiac surgery. Anesth Analg 1992;74:219–25.
28. Konrad CJ, Markl TJ, Schuepfer GK, et al. In vitro effects of different medium molecular hydroxyethyl starch solutions and lactated Ringer’s solution on coagulation using SONOCLOT. Anesth Analg 2000;90:274–9.
29. Treib J, Haass A, Pindur G, et al. All medium starches are not the same. Transfusion 1996;36:450–5.
© 2006 International Anesthesia Research Society
30. Dietrich GV, Schueck R, Menges T, et al. Comparison of four methods for the determination of platelet function in whole blood in cardiac surgery. Thromb Res 1998;89:295–301.