Hydroxyethyl starch (HES) solutions have largely replaced human albumin and plasma as colloidal fluids for modest intravascular volume expansion. Nevertheless, many clinicians are concerned that HES may increase the risk of bleeding in the perioperative setting (1). Clinical investigations have demonstrated that the “first generation” high molecular weight HES 450/0.7 (Hetastarch®; mean molecular weight [Mw], 450 kDa, molar substitution [MS] 0.7) causes significant abnormalities of some laboratory tests of hemostasis (2,3) by decreasing factor VIII activity and von Willebrand factor (vWF) (4), inhibiting platelet function (5–7), and exerting unfavorable effects on fibrin clots and fibrinolysis (8,9). HES preparations that have lower Mw and less MS appear to have fewer deleterious effects on coagulation (1).
In Europe, a variety of second and third generation (medium and low Mw, low MS) HES are widely popular (10). In the United States, Hespan® and Hextend® (Mw 550 kDa; MS 0.7; suspended in physiologically balanced electrolytes) are the only HES approved for perioperative intravascular volume expansion, whereas, in Canada HES 264/0.45 (Pentaspan®; Bristol-Myers Squibb, St-Laurent, Quebec) usage predominates. Hespan® and Hextend® contain a high Mw, slowly metabolized HES whereas HES 264/0.45 is a medium Mw (Mw 264 kDa, MS 0.45), rapidly degradable HES (5). Pentastarches overcome many of the shortcomings of higher Mw HES preparations; they have a short plasma half-life and may have fewer detrimental effects on coagulation than HES 450/0.7 (11,12). Current guidelines recommend limiting the maximum volume of HES 264/0.45 to 2 L or 28 mL/kg within a 24-h period, based on the concerns regarding coagulopathy that existed with previous HES preparations (Pentaspan® product monograph). Clinical experience at our institutions suggests that HES 264/0.45 may be administered in volumes exceeding this limit without specific coagulation effects. However, this has never been rigorously tested under clinical conditions.
A comparison of HES 264/0.45 and HES 450/0.7 in healthy volunteers not undergoing surgery showed that these HES preparations have similar intravascular volume-expanding effects, although HES 264/0.45 caused fewer hemostatic abnormalities (12). One study has investigated the effects of large doses of pentastarch in a surgical setting (13). It demonstrated that HES 200/0.5 in doses up to 3 L/24 h (180 g/24 h) does not appear to affect coagulation to a larger extent than similar volumes of 5% human albumin.
To investigate the effect of large doses of HES 264/0.45 further, we examined the effects of HES 264/0.45 and 5% albumin (up to 45 mL/kg per 24 h) on coagulation and physiological variables of patients undergoing major head and neck ablative surgery with microvascular reconstruction. This surgical procedure was chosen because large volumes of IV fluid are routinely administered during the long perioperative course whereas blood loss and iatrogenic disorders of coagulation are modest, making it an excellent in vivo model to study the effects of large-volume IV therapy (14). The primary hypothesis of this study is that subjects receiving HES 264/0.45 (up to 45 mL/kg per 24 h) for intravascular volume expansion will not have clinically significantly different laboratory indices of coagulation compared with subjects receiving 5% human albumin.
The research ethics boards of the two participating hospitals approved the study protocol and all subjects provided written informed consent before participating in the trial.
Adult patients undergoing surgical ablation of oro-pharyngeal cancer with free-flap reconstruction were eligible to enter the study. Exclusion criteria were ASA physical status classification III–IV, cardiac insufficiency (New York Heart Association class III and IV), pancreatitis (serum amylase >400 U/L), severe hepatic dysfunction (aspartate aminotransferase >50 U/L, alanine aminotransferase >60 U/L), renal dysfunction (serum creatinine >140 μmol/L or >1.58 mg/dL), anemia (hemoglobin <110 g/L), coagulation abnormalities (international normalized ratio [INR] >1.2, activated partial thromboplastin time [aPTT] >40 s, platelet count <120 × 109/L, fibrinogen <1.5 g/L), ingestion of nonsteroidal antiinflammatory drugs or acetylsalicylic acid within 10 days of surgery, and previous major head and neck surgery with free-flap reconstruction.
The subjects were randomly allocated to the two study groups at induction of anesthesia using a computerized randomization schedule. Randomization was stratified within each participating hospital and subjects were allocated in blocks of 4–6 patients.
A nurse placed the study colloids in a masked and locked container and primed the delivery pump and tubing, which was covered with opaque tape to maintain the blinding. The biostatistician who prepared the randomization schedule and the nurse who prepared the study colloids did not participate in other aspects of the trial. All subjects, health care workers involved in patient care, and other study personnel were blinded as to subject allocation. The allocation groups remained blinded until after all data were analyzed.
Subjects received either HES 264/0.45 or 5% albumin. The study colloids were administered via infusion pumps to a maximum of 45 mL/kg per 24 h. All IV fluids and blood products were administered via infusion pumps according to predetermined study protocols. Calculated preoperative fluid deficit and intraoperative insensible losses were replaced intraoperatively with lactated Ringer’s solution. Study colloid was administered in boluses of 250 mL over 20 min to keep central venous pressure (CVP) between 7 and 10 mm Hg. Packed red blood cells were administered to maintain hemoglobin >80 g/L. Fresh-frozen plasma was infused to maintain the INR <2.0. Platelets were administered for platelet count <50 × 109/L. Study subjects did not receive antithrombotic prophylaxis during the study period.
Subjects did not receive sedative premedication. Anesthesia was induced with fentanyl 2–4 μg/kg, thiopental 3–5 mg/kg, and succinylcholine 1.5 mg/kg, followed by endotracheal intubation. Anesthesia was maintained with N2O/O2 (60%/40%), isoflurane (0.5%–1.5%) and intermittent doses of morphine 0.1–0.2 mg/kg or fentanyl 1.5 –2.0 μg/kg. Paralysis was maintained with pancuronium. Normothermia was maintained by means of a warming blanket, IV fluid warmer, and warming of ambient air. Perioperative monitoring included continuous arterial blood pressure (radial arterial catheter), continuous CVP (subclavian vein catheter), electrocardiogram, pulse oximeter, end-tidal gas measurements, rectal temperature, and urinary output.
Blood samples were obtained from the arterial catheter shortly after induction of anesthesia (baseline) and after subjects received 15 mL/kg, 30 mL/kg, and 45 mL/kg of study colloid. Hematocrit, platelet count, INR, aPTT, thrombin time, fibrinogen activity, factor VIII activity level, vWF antigen level (vWFL), vWF activity (vWFA), d-dimer level, plasminogen level, α-2-antiplasmin level, and antithrombin III level were measured according to standard laboratory techniques.
All data were analyzed using intention-to-treat principles. The primary outcome measures were the aPTT and INR. These variables were chosen because documented effects of HES on factor VIII and fibrinogen are reflected in the aPTT. Secondary outcomes included other laboratory tests of coagulation (fibrinogen, factor VIII, vWFA, vWFL, and platelet concentration) and clinical indices of coagulation (transfusion of allogeneic blood, fresh-frozen plasma, and platelets), and efficacy of albumin and HES 264/0.45 as intravascular volume expanders (amount of study colloid infused during the study period to maintain CVP 7–10 mm Hg). Patients with group O blood have lower levels of VIII/vWF complex and may be more susceptible to deleterious effects of HES than others (15). Therefore, we performed a post hoc analysis to investigate the effects of HES 264/0.45 on factor VIII and vWFL in blood group O subjects. A sample size calculation performed a priori determined that 68 subjects overall would be needed to demonstrate a difference in the aPTT of 1.4 s between the 2 treatment groups with a power of 90% and two-sided α = 0.05.
Demographic factors and coagulation variables are described using appropriate summary statistics such as means and standard deviations for continuous variables and frequencies for discrete variables. To test the differences between the groups with respect to these factors, we used Student’s t-tests for the continuous variables whose distribution was normal; otherwise we used the Wilcoxon-Mann-Whitney U-test. For the discrete variables, the differences between groups were assessed using the χ2 test or the Fisher’s exact test if the cell counts were not sufficiently large enough for χ2 analysis. The coagulation variables at baseline, 15 mL/kg, 30 mL/kg, and 45 mL/kg were analyzed using repeated-measures analysis for volume and group and interaction between volume and group and adjusting these for sex, weight, and the time from baseline at which the volumes were infused. The covariance structures or random coefficient models were chosen based on the variogram and the AIC/BIC criteria for optimal fit. The residuals were inspected to check for any deviations from the model assumptions. P values are for two-sided tests and significance was declared for P < 0.05. All analyses were performed using SAS v. 8.02 (SAS, Cary, NC).
One-hundred-thirteen patients were assessed for eligibility in the trial, 63 were excluded and 50 were randomized to one of the treatment allocations (Fig. 1). Sixteen subjects allocated to the albumin group and 18 subjects allocated to the HES 264/0.45 group received the full volume of 45 mL/kg over 24 h (Fig. 1). The study was closed before reaching the target sample size of 68 subjects because the recruitment rate became unacceptably small after 4 yr. The decision to close the study was made without knowledge of treatment specific outcomes. Two patients did not complete the study because the planned surgical procedure was abandoned intraoperatively for unrelated reasons. Data from 24 subjects in each study group were analyzed. Both groups were similar with respect to age, ASA status, blood groups, and types of procedures performed (Table 1). There was a larger percentage of women in the HES 264/0.45 group; this accounted for significant differences in weight and height between groups.
Both albumin and HES 264/0.45 effectively maintained physiologic variables (CVP, mean arterial blood pressure [MAP], and heart rate [HR]) in the perioperative and postoperative periods (Table 2). Inconsistent intergroup differences occurred, but they never exceeded physiological limits. A significant difference was observed in the mean infused volume of study colloid administered to both groups (Table 3). Subjects in the HES 264/0.45 group received on average 20 mL/kg more crystalloid than the albumin group during the study period (P = 0.12).
The aPTT and INR increased significantly with increasing volumes of colloid and crystalloid infused in both groups. There was a significant difference in the aPTT between the two study groups after the infusion of 30 mL/kg and 45 mL/kg (Fig. 2, panel A) and for INR after infusion of 15, 30 and 45 mL/kg (Fig. 2, panel B). vWFA, vWFL, and factor VIII in subjects administered albumin remained unchanged with the infusion of up to 45 mL/kg; however, there was a significant dose-dependent reduction in vWFA, vWFL and factor VIII in those receiving HES 264/0.45 (Fig. 3). Thrombin time, fibrinogen, d-dimer, plasminogen, α-2-antiplasmin, and antithrombin III levels were not significantly different between the two groups (data not shown).
The mean baseline vWFL for blood group O patients and nongroup O patients were 1.20 ± 0.47 U/mL and 1.36 ± 0.39 U/mL respectively (not significant). Baseline factor VIII levels were significantly less in group O patients (group O, 1.16 ± 0.47 U/mL; nongroup O, 1.58 ± 0.53 U/mL; P = 0.008). There was no significant difference in the maximum decrease in vWFL (group O, 0.67 ± 0.57 U/mL; nongroup O, 0.63 ± 0.32 U/mL; P = 0.93) or factor VIII (group O, 0.75 ± 0.29 U/mL; nongroup O, 1.18 ± 0.59 U/mL; P = 0.42).
Nineteen percent of the subjects in the albumin group (n = 5) and 54% in the HES 264/0.45 group (n = 13) received allogeneic blood transfusions (P = 0.015). Subjects in the albumin group received a median of 1 U each compared with 3 U in the HES 264/0.45 group.
There were three patients with potentially serious adverse events. One patient in the HES 264/0.45 group had postoperative bleeding that required re-exploration. On re-exploration a surgical cause for the bleeding was found. This patient received 5 U of packed red blood cells, 3 units of fresh-frozen plasma, and 5 U of platelets. One patient in the HES 264/0.45 group developed a pulmonary embolus 72 h after surgery. The patient was treated with IV heparin and the condition resolved. One patient in the albumin group developed pulmonary edema in the recovery room. This patient required an additional 2 h of ventilatory support immediately postoperatively.
In this randomized trial, we demonstrated that infusion of HES 264/0.45 or 5% albumin in volumes up to 45 mL/kg over 24 hours in patients undergoing major reconstructive head and neck surgery resulted in larger dose-dependent increases in aPTT and INR in subjects receiving HES 264/0.45. These changes were associated with HES 264/0.45 mediated decreases in vWFA, vWFL, and factor VIII level. The frequency of allogeneic red blood cell transfusion and the median number of red cell transfusions were also increased in subjects allocated to HES 264/0.45.
This study using HES 264/0.45 confirms previous studies in healthy volunteers demonstrating that pentastarches (HES 200/0.5 and HES 265/0.45) reduce factor VIII activity, vWFL, and vWFA (12,16,17). Our results contrast the results of Vogt et al. (13), who compared HES 200/0.5 to 5% albumin (up to 3000 mL) in patients undergoing total hip arthroplasty. Although the Quick’s prothrombin value was transiently increased in the HES 200/0.5 group, there were no between-group differences in PTT, blood loss, or administered blood components. Two possibilities may explain our discrepant findings. First, in the Vogt et al. study, only 50% (n = 10) and 28% (n = 6) of the subjects allocated to HES 200/0.5 and albumin, respectively, received 3000 mL (fewer than 15 subjects received 2000 mL in either study group). Therefore, their sample size may have been inadequate to identify significant effects on coagulation, blood loss, or blood product use. Alternatively, differences in the surgical populations we studied may have affected the results. Liberation of bone and fat debris during major orthopedic surgery activates coagulation (18). Thus, patients undergoing major orthopedic procedures may compensate more adequately for the effects of HES on coagulation (17).
Both study colloids were able to maintain CVP, HR, and MAP in the perioperative period within physiologically normal variables. However, the HES 264/0.45 group received approximately 15% more colloid. This suggests that albumin is at least as effective as HES 264/0.45 in augmenting CVP in this clinical setting despite the fact that albumin was infused as a 5% solution and HES 264/0.45 as a 10% solution.
Our subgroup analysis confirmed results from Huraux et al. (15) that patients with blood type O have lower factor VIII levels than other blood groups. However, we could not corroborate their observation that patients with blood type O are particularly vulnerable to the effects of HES. The small number of subjects in the subgroup we analyzed may explain this discrepancy.
We chose to study subjects undergoing elective major reconstructive surgery because large volumes of colloids are routinely administered to improve perfusion to the transferred free flap and significant coagulation abnormalities are seldom encountered (14,19). In studying this patient population, we overcame some of the limitations of previous studies that have investigated the hemostatic effects of HES 264/0.45 (20). These studies were either done in vitro (21), in healthy volunteers (12,22), or in cardiac surgery patients (23). It is not possible to reproduce surgical activation of coagulation in in vitro studies or in healthy volunteers (17). In cardiac surgery, the large doses of heparin administered and the effects of the cardiopulmonary bypass circuit alter coagulation, platelet function, and fibrinolysis. In the current study, we chose to administer the study colloids guided by the CVP with the aim of increasing venous return (intravascular volume expansion) to maintain adequate tissue perfusion. This approach to fluid management parallels clinical routine and facilitates the interpretation of our results in the context of routine clinical care. Several study design features reduce the likelihood of bias. The triple-blind design and strict infusion algorithms ensured that physician bias did not influence the decision to infuse colloids, crystalloids, or blood products.
Our results support the common practice to switch to alternative colloids once approximately 30 mL/kg per 24 hours of HES 264/0.45 (or the closely related HES 200/0.5) has been administered (10,24). At the dose of 30 mL/kg over 24 hours, there was a significant difference in the INR and aPTT between the two groups; nevertheless, the increased INR (1.47) and aPTT (47 s) in the HES 264/0.45 group did not necessarily predict increased risk for surgical bleeding. At the dose of 45 mL/kg, the INR level in the HES 264/0.45 group was increased to 1.7 and the aPTT to 56 s. In some clinical circumstances this will be well tolerated by patients without untoward events. However, in the setting of major surgery with continuous blood loss, or in circumstances where postoperative hematoma may have devastating consequences, many physicians will administer fresh-frozen plasma prophylactically. Furthermore, there are no data on the risk of bleeding when large doses of HES are administered to patients with thrombocytopenia, vitamin K deficiency, or a preexisting clotting disorder, and caution is advised under these circumstances. The 50% reductions in vWFL, vWFA, and factor VIII activity after HES 264/0.45 infusions of >30 mL/kg is of further concern, as data from hemophiliacs undergoing surgery suggest that the risk of surgical bleeding increases when levels of these clotting factors are <50% of normal values (25). Conroy et al. (26) have demonstrated that desmopressin (DDAVP) may overcome the decrease in factor VIII caused by HES. They suggest that DDAVP should be considered a choice of treatment in clinical situations where patients have received HES and bleeding occurs.
There are several limitations to this study. ASA class III and IV patients were excluded from participation. We excluded this group for ethical concerns surrounding the possibility of complications from excessive intravascular volume expansion. Thus, the findings in our study should not be extrapolated to ASA class III–IV patients. Our sample size calculation determined that we would need to recruit 68 subjects to obtain a power of 0.90. As we recruited only 48 subjects, the study had a calculated power of 0.77. The effects of pentastarches (HES 264/0.45 and HES 200/0.5) on platelet function are controversial, with some studies showing minimal effects (12) and others demonstrating platelet inhibition (6,7). We did not measure thromboelastography or platelet function in the current study, and, therefore, our results do not evaluate overall impairment of the hemostatic process or platelet function but focus on impairments in the protein coagulation cascade that produce fibrin. Assessment of platelet function would have been helpful in delineating further the effect of large doses of HES 264/0.45 on clot formation.
Despite the use of a randomized computer-generated protocol, more women were allocated to the HES 264/0.45 group than the albumin group. Gender, however, does not have a clinically significant effect on coagulation variables and, therefore, the unequal number of women in each group did not likely influence the observed differences in coagulation factor levels or activities (27). Conversely, the more frequent transfusions in the HES 264/0.45 group may be partially explained by the larger number of women allocated to this group, as previous studies have shown that gender may be an independent predictor of the need for transfusion in major surgery (28). Our study was not designed to measure allogeneic transfusion as a primary outcome. Therefore, the association we observed between increased allogeneic red blood cell transfusion and HES 264/0.45 must be confirmed by a larger adequately powered trial with allogeneic blood transfusion as a primary outcome before definitive conclusions on this association may be made.
We conclude that in patients undergoing major head and neck surgery, large volumes of HES 264/0.5 may have a less favorable coagulation profile and may lead to increased allogeneic blood transfusion compared with 5% albumin. Additional large scale, randomized, controlled studies with allogeneic blood transfusion as a primary outcome are needed to confirm these results.
We would like to thank Jo Carroll for the preparation of blinded colloid solutions used in this study.
1. Boldt J. New light on intravascular volume replacement regimens: what did we learn from the past three years? Anesth Analg 2003;97:1595–604.
2. Martin G, Bennett-Guerrero E, Wakeling H, et al. A prospective, randomized comparison of thromboelastographic coagulation profile in patients receiving lactated Ringer’s solution, 6% hetastarch in a balanced-saline vehicle, or 6% hetastarch in saline during major surgery. J Cardiothorac Vasc Anesth 2002;16:441–6.
3. Knutson JE, Deering JA, Hall FW, et al. Does intraoperative hetastarch administration increase blood loss and transfusion requirements after cardiac surgery? Anesth Analg 2000;90:801–7.
4. Stump DC, Strauss RG, Henriksen RA, et al. Effects of hydroxyethyl starch on blood coagulation, particularly factor VIII. Transfusion 1985;25:349–54.
5. Treib J, Baron JF, Grauer MT, Strauss RG. An international view of hydroxyethyl starches. Intensive Care Med 1999;25:258–68.
6. Franz A, Braunlich P, Gamsjager T, et al. The effects of hydroxyethyl starches of varying molecular weights on platelet function. Anesth Analg 2001;92:1402–7.
7. Stogermuller B, Stark J, Willschke H, et al. The effect of hydroxyethyl starch 200 kD on platelet function. Anesth Analg 2000;91:823–7.
8. Strauss RG, Stump DC, Henriksen RA, Saunders R. Effects of hydroxyethyl starch on fibrinogen, fibrin clot formation, and fibrinolysis. Transfusion 1985;25:230–4.
9. Carr ME Jr. Effect of hydroxyethyl starch on the structure of thrombin- and reptilase-induced fibrin gels. J Lab Clin Med 1986;108:556–61.
10. Langeron O, Doelberg M, Ang ET, et al. Voluven, a lower substituted novel hydroxyethyl starch (HES 130/0.4), causes fewer effects on coagulation in major orthopedic surgery than HES 200/0.5. Anesth Analg 2001;92:855–62.
11. Strauss RG, Stansfield C, Henriksen RA, Villhauer PJ. Pentastarch may cause fewer effects on coagulation than hetastarch. Transfusion 1988;28:257–60.
12. Strauss RG, Pennell BJ, Stump DC. A randomized, blinded trial comparing the hemostatic effects of pentastarch versus hetastarch. Transfusion 2002;42:27–36.
13. Vogt NH, Bothner U, Lerch G, et al. Large-dose administration of 6% hydroxyethyl starch 200/0.5 total hip arthroplasty: plasma homeostasis, hemostasis, and renal function compared to use of 5% human albumin. Anesth Analg 1996;83:262–8.
14. Sigurdsson GH. Perioperative fluid management in microvascular surgery. J Reconstr Microsurg 1995;11:57–65.
15. Huraux C, Ankri AA, Eyraud D, et al. Hemostatic changes in patients receiving hydroxyethyl starch: the influence of ABO blood group. Anesth Analg 2001;92:1396–401.
16. de Jonge E, Levi M, Buller HR, et al. Decreased circulating levels of von Willebrand factor after intravenous administration of a rapidly degradable hydroxyethyl starch (HES 200/0.5/6) in healthy human subjects. Intensive Care Med 2001;27:1825–9.
17. Jamnicki M, Bombeli T, Seifert B, et al. Low- and medium-molecular-weight hydroxyethyl starches: comparison of their effect on blood coagulation. Anesthesiology 2000;93:1231–7.
18. Hobisch-Hagen P, Wirleitner B, Mair J, et al. Consequences of acute normovolaemic haemodilution on haemostasis during major orthopaedic surgery. Br J Anaesth 1999;82:503–9.
19. Khouri RK. Avoiding free flap failure. Clin Plast Surg 1992;19:773–81.
20. Haisch G, Boldt J, Krebs C, et al. The influence of intravascular volume therapy with a new hydroxyethyl starch preparation (6% HES 130/0.4) on coagulation in patients undergoing major abdominal surgery. Anesth Analg 2001;92:565–71.
21. 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.
22. Ruttmann TG, James MF, Aronson I. In vivo
investigation into the effects of haemodilution with hydroxyethyl starch (200/0.5) and normal saline on coagulation. Br J Anaesth 1998;80:612–6.
23. Boldt J, Zickmann B, Herold C, et al. Influence of hypertonic volume replacement on the microcirculation in cardiac surgery. Br J Anaesth 1991;67:595–602.
24. Neff TA, Doelberg M, Jungheinrich C, et al. Repetitive large-dose infusion of the novel hydroxyethyl starch 130/0.4 in patients with severe head injury. Anesth Analg 2003;96:1453–9.
25. Cohen AJ, Kessler CM, Ewenstein BM. Management of von Willebrand disease: a survey on current clinical practice from the haemophilia centres of North America. Haemophilia 2001;7:235–41.
26. Conroy JM, Fishman RL, Reeves ST, et al. The effects of desmopressin and 6% hydroxyethyl starch on factor VIII:C. Anesth Analg 1996;83:804–7.
27. Kain K, Carter AM, Bamford JM, et al. Gender differences in coagulation and fibrinolysis in white subjects with acute ischemic stroke. J Thromb Haemost 2003;1:390–2.
© 2005 International Anesthesia Research Society
28. Karkouti K, Cohen MM, McCluskey SA, Sher GD. A multivariable model for predicting the need for blood transfusion in patients undergoing first-time elective coronary bypass graft surgery. Transfusion 2001;41:1193–203.