Portal de Periódicos da CAPES

Hydroxyethyl Starch

1997; Lippincott Williams & Wilkins; Volume: 84; Issue: 1 Linguagem: Inglês

10.1213/00000539-199701000-00037

1526-7598

Brian B. Warren, Marcel E. Durieux,

Hemostasis and retained surgical items

Hydroxyethyl starch (HES) and albumin are the two colloids administered most frequently for intravascular volume expansion during the perioperative period. Ideally, the selection between the two would be based only on relative risks and benefits in each individual patient. However, economic issues have become increasingly important in choosing between medical therapies. In terms of cost-containment, HES is an attractive choice, because it is considerably less expensive than albumin. As an example, at the University of Virginia Medical Center, the hospital cost of 500 mL (1 unit) HES is $43, whereas 500 mL albumin costs $66. Between March 1994 and March 1995, 536 L HES and 422.5 L albumin were used in the operating rooms. Assuming that HES use is similar among hospitals, extrapolation to all hospitals in the United States suggests that exclusive use of HES could save approximately $50 million/year. However, the economic benefit of using HES instead of albumin as a colloid for volume expansion must be weighed against potential adverse effects. Particularly, effects of HES on the coagulation system have received attention [1-18]. This study will give an overview of the physical chemistry and pharmacology of HES, and review data available on adverse effects to allow clinicians to make an informed choice as to which colloid to use. The first colloid used clinically for intravascular volume expansion was acacia, a mixed polysaccharide used during World War I. Several colloids have followed (including gelatins, dextran, and synthetic polypeptides), but all of these occasionally induced anaphylactoid reactions or coagulation defects. Some were eliminated too rapidly from the circulation, whereas others remained in the body indefinitely. In 1963, Thompson and Walton suggested HES as an alternative.1 (1) Thompson WL, Walton RF. Blood changes, renal function and tissue storage following massive infusion of hydroxyethyl starches [abstract]. Fed Proc 1963;22:640. Physical Chemistry Structure HES is synthesized from amylopectin, a waxy starch derived from maize or sorghum. It consists of D-glucose units linked in a branching structure (Figure 1). The number of alpha 1-6 branch points can vary, and this characteristic is expressed as the degree of branching: the ratio of branch points to glucose units. A reaction between ethylene oxide and amylopectin in the presence of an alkaline catalyst attaches hydroxyethyl groups to carbons 2, 3, or 6 of the glucose moieties. These hydroxyethyl groups retard hydrolysis of the compound by amylase, thereby delaying its breakdown and elimination from the blood. The degree of substitution (expressed as a number between 0 and 1) indicates the fraction of glucose moieties bearing a hydroxyethyl group. It can be controlled by varying the reaction duration, whereas the size of the molecules can be modified by acid hydrolysis of the parent compound.Figure 1: A segment of hydroxyethyl starch with hydroxyethyl groups indicated (CH (2) CH2 OH). These can be attached at any free hydroxyl group.As an example, a frequently used form of HES (Hespan; DuPont Pharma, Wilmington, DE) has a degree of substitution of 0.7 (7 hydroxyethyl groups/10 glucose units) and a degree of branching of approximately 1:20 (one alpha 1-6 branch/20 glucose units). The solution as prepared for clinical use consists of 6 g of HES/100 mL of normal saline, with an osmolarity of 310 mOsm/L and a pH of 3.5-7.0. Molecular Weight Depending on the distribution of molecular weights of the constituting particles, colloids can be either monodisperse or polydisperse. A monodisperse colloid consists of molecules of one molecular weight only, whereas a polydisperse colloid contains a range of weights. For example, albumin is a monodisperse colloid with a molecular weight of 69 kDa [19]. For polydisperse colloids, molecular weight can be reported either as number average molecular weight (Mn; sample mass in grams divided by the total number of molecules) or as weight average molecular weight (Mw; the sum of each molecule's weight divided by the sample weight times the weight of a molecule) [19,20]. These numbers can be quite different; for example, HES has a Mw of 480 kDa (with 80% of the polymers between 30 and 2400 kDa), whereas its Mn is 60-80 kDa [19]. Several HES fractionation products are under study, because they may have the ability to seal selectively the endothelial "pores" or gaps that develop in microvessels after different forms of injury [21,22]. Therefore, they might prevent leakage of plasma protein (primarily albumin) from the intravascular to the extravascular space, which will induce osmotic water loss from the intravascular space and lead to fluid extravasation. Examples of these compounds are pentastarch (Mw 280; Mn 120; range 10-1000 kDa) and pentafraction (Mw 264; Mn 63; range 50-1000 kDa). Both compounds have a substitution ratio of 0.5, lower than HES. Figure 2 shows the molecular weight distribution of various forms of HES.Figure 2: Molecular weight distributions of various forms of hydroxyethyl starch (HES). Bar indicates the weight average molecular weight, the lines indicate the range of values. Albumin, a monodisperse colloid, is included for comparison.Pharmacology HES is removed from the circulation by two major mechanisms: renal excretion and redistribution. A third pathway, elimination through the gastrointestinal tract, is of minimal importance. The predominant mechanism, renal excretion, consists of two phases. The first occurs almost immediately upon administration, because polymers with a molecular weight less than 59 kDa are eliminated rapidly by glomerular filtration. A second phase of glomerular filtration is more prolonged and occurs as HES molecules are metabolized. As previously noted, the presence of hydroxyethyl groups on amylopectin slows enzymatic degradation of HES. Nonetheless, hydrolysis by amylase does occur, and serum amylase activity actually increases after HES administration. Once the product of amylase digestion is smaller than 72 kDa, it can be excreted renally. The second mechanism of removal from the circulation is redistribution: the uptake and temporary storage of HES in the tissues. After 24 hours, 23% of the total dose is extravasated into the interstitial space [19,20,23,24]. As a result of these processes, only 38% of the initial dose remains intravascular 24 hours after administration, whereas 39% of the product has appeared in the urine and 23% is sequestered extravascularly. Trace amounts of HES are detectable in the circulation for up to 17-26 weeks. The duration of volume expansion with HES is approximately 24 hours, with 29%-38% of the colloid still available intravascularly after this period [19,20,23,24]. A similar duration of volume expansion has been reported for albumin [19]. For purposes of comparison, a few comments about albumin are in order. Albumin accounts for 70%-80% of the plasma oncotic pressure. The remainder of osmotic pressure is produced by other plasma proteins with a collective molecular weight of 150 kDa. The total body albumin stores are 4.5-5.0 g/kg, one third of which is intravascular. Five percent of albumin leaks from the circulation per hour, whereas 90% of the extravascular pool returns to the circulation each day. Synthesis of albumin occurs in hepatocytes, and the compound is catabolized at a rate of 10% daily, primarily in the reticuloendothelial system [19]. HES and albumin expand intravascular volume equally effectively. For example, Lazgrove et al. [26] studied 10 acutely ill postoperative patients of reasonable hemodynamic stability and observed similar hemodynamic responses to infusion of HES or albumin. However, although HES and albumin have similar volume-expanding properties, this should not suggest that the compounds are similar in all other regards. Albumin acts as an important carrier protein for many drugs relevant to anesthesiologists. In addition, it can be used as a marker of nutritional status and may have other physiologic functions. Although controversial, low serum albumin levels might be an indication for selecting albumin, rather than HES, as a volume expander. Two circumstances may preclude the use of albumin. The first is availability; this depends on the donor supply, which may be low at times. The second is in the treatment of Jehovah's Witnesses, some of whom, for religious reasons, will not receive albumin. In both instances, HES may be an appropriate substitute. Adverse Effects HES has been associated with several adverse effects. The most common of these will be described in this section, whereas effects on coagulation will be discussed in the next section. Anaphylactoid Reactions Whereas anaphylactic reactions are mediated by antigen-antibody interactions in subjects previously exposed to a drug, anaphylactoid reactions (although often clinically indistinguishable) are considered caused by direct effects of the drug on mast cells and basophils, with resultant release of histamine and other mediators. Ring and Messmer [26] studied 16,405 patients who received HES, and observed 14 anaphylactoid reactions, an incidence of 0.085% (1 in 1200). These reactions were typically mild to moderate, and only one was classified as severe. In two case reports of moderate intraoperative reactions to HES (hypotension, respiratory disturbance, and rash), the patients were treated with steroids, cimetidine, and diphenhydramine, as well as increased inspired oxygen concentrations and ephedrine for circulatory support [27]. The risk of anaphylactoid reactions after albumin administration is 0.011% [26], or approximately one eighth the risk associated with HES. Serum Amylase A serum macroamylasemia after HES administration has been described by Kohler et al. [28]. Serum amylase increased by an average of 201 U/L; its renal excretion decreased regardless of the degree of preexisting renal disease. However, the production of amylase by the pancreas did not change, and no damage to the pancreas was noted. The decreased renal elimination of amylase resulted from formation of a complex of higher molecular weight HES molecules and amylase. Glomerular filtration of amylase decreased, with a resultant increase in serum amylase levels. Serum amylase remained markedly elevated for 72 hours and therefore should not be used as a diagnostic marker of pancreatic disease for a period of three to five days after HES administration [28]. Pruritus Severe pruritus has been associated with the administration of HES. Tissue redistribution of HES occurs, with storage of the compound in the mononuclear phagocyte system of various organs; deposits in the skin induce pruritus. Because this is not a result of histamine release [29], antihistamines are ineffective; but topical capsaicin (available over the counter) has been reported to be useful [22]. Other Although adverse reactions other than those described herein are mentioned in manufacturer's documentation (vomiting, fever, chills, parotid enlargement, headaches, myalgias, and pulmonary edema), these have not been reported in the literature. Some of these may have resulted from mild anaphylactoid reactions or fluid overload. Because HES is not derived from human blood, it does not represent a risk of disease transmission. Albumin, in contrast to HES, is derived from human blood. Therefore, the compound is heat-pasteurized (60 degrees C for 10 hours) to eradicate blood-borne pathogens, and the donor plasma is screened for human immunodeficiency virus and hepatitis B and C. A MEDLINE review of the past 30 years revealed no reported cases of hepatitis or human immunodeficiency virus transmitted by albumin. In rare instances, albumin may contain contaminants such as aluminum, nickel, and chromium; accumulation of these metals can occur in patients with impaired renal function [30]. Such contamination has not been reported for HES. HES and Coagulation HES can affect the coagulation process [1-14,31-34]. If clinically significant, this adverse effect would have enormous importance. Decreased perioperative hemostasis or an induced coagulopathy can have serious consequences for the patient, in addition to sizable economic impact. Albumin, in contrast, does not seem to affect coagulation. Johnson et al. [35] demonstrated a coagulopathy in patients in traumatic hemorrhagic shock treated with the compound. However, each patient received as much as 3 L (150 g) of albumin/day for three to five days. This massive infusion volume diminishes the clinical relevance of this study. A review of the MEDLINE database revealed no other case reports of coagulopathy after albumin administration. A literature review of the past 30 years on the effects of HES on coagulation revealed controversial and conflicting data. Eighteen papers were available, and these were classified into three groups according to the coagulation effects observed. The first group demonstrated no effect of HES on coagulation, the second group found changes in laboratory test results only, and the third group demonstrated a clinically significant effect on coagulation. Each group will be discussed separately (Table 1).Table 1: Studies of the Effect of Hydroxyethyl Starch on Coagulation, Grouped by ResultNo Effects Six studies found no effect of HES on coagulation. Two of these compared HES with albumin in 40-60 postoperative cardiac surgery patients. These studies revealed no change in coagulation variables, such as prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen, Factor VIII, von Willebrand factor, platelet function, or chest tube drainage [36,37]. In these studies, the compound was administered postoperatively. Two other studies compared, in 5-20 patients, the coagulation effects of HES with those of albumin when administered intraoperatively; no change in any of the studied coagulation variables was noted [38,39]. In a small group of 12 patients treated for septic shock, the coagulation effects of HES and albumin were compared; again, no differences were found [40]. Lastly, Gold et al. [31] randomly allocated 40 patients undergoing abdominal aortic aneurysm repair to receive either HES or albumin (1 g/kg) intraoperatively. No difference in PTT, activated clotting time, PT, platelet count, or bleeding time were found. Also, there was no difference in estimated blood loss or amount of packed red blood cells administered. Because of its intraoperative, randomized design and larger group sizes, this study may be more relevant than the previous ones. Effects on Laboratory Values Only Most of the studies reviewed find that HES affects coagulation. In studies of healthy volunteers, decreases in Factor VIII, fibrinogen, and von Willebrand factor, and increases in PTT, PT, and bleeding time were noted repeatedly after dosages of 500-1000 mL HES [4,5,7,34], and the same changes were seen in an animal model [6]. Alexander et al. [33] found a decrease in Factor VIII and fibrinogen, although the amount of change varied widely between subjects and did not correlate with the amount of HES administered. Stump et al. [3] demonstrated decreases in Factor VIII and von Willebrand factor, and increases in PTT, although again the changes did not correlate with the plasma level of HES. No bleeding complications were noted, but sample sizes were small (10-30 subjects), and no surgery took place. Of two small series of surgical patients in which one of the groups received at least 500 mL of HES, one revealed decreases in Factor VIII, fibrinogen, and platelet aggregation [1], whereas the other showed decreases in Factors VII and IX and in platelet aggregation [16]. No bleeding tendency was noted. Finally, when 20 mL/kg of HES was added to the cardiopulmonary bypass priming solution of patients undergoing cardiac surgery, decreases in Factor VIII and von Willebrand factor levels and increases in PTT were seen [2]. Although few studies are available, it appears that pentastarch has fewer effects on coagulation than HES. When compared with HES, pentastarch had little effect on Factor VIII, shortened thrombin times to a significantly lesser degree, exerted no effect on urokinase-activated clot lysis time, and did not prolong the bleeding time [34]. Clinically Significant Effects Studies in cardiac surgery patients have found coagulopathies or other adverse effects after HES administration. Boldt et al. [8] randomly allocated 75 patients undergoing cardiopulmonary bypass to receive either albumin, gelatin, or one of two HES fractions: Mw 450 kDa or Mw 200 kDa. In the 15 patients who received an average of 840 mL of Mw 450 kDa HES, an increase in chest tube drainage was noted on the first postoperative day. Although not statistically significant, these patients also tended to require more units of packed red blood cells (Table 2). The platelet count did not change, but platelet aggregation decreased in this group and correlated with blood loss. In a cohort and case-control study by Villarino et al. [9], management of a series of cardiac surgery patients was reviewed after an outbreak of bleeding was noted during a two-month period (28 operations). During this period, the average volume of HES used intraoperatively was 19 mL/kg in 6 of the 28 patients, compared with 14 mL/kg in the other 22 patients. These six patients had increased PT and PTT values, compared with the others and received significantly more blood products: 12 U vs 2 U packed red blood cells, 23 U vs 2.5 U platelets, and 11 U vs 1 U fresh frozen plasma. Also, five of these six patients required reexploration for "nonsurgical bleeding," opposed to 5 of 440 previous patients.Table 2: Chest Tube Drainage and Units of Packed Red Blood Cells Transfused in Cardiac Surgery Patients [8]Cope and Tribble [41] retrospectively analyzed data on 127 consecutive coronary artery bypass patients who received an average of approximately 800 mL of HES either intraoperatively after bypass or during the first eight postoperative hours. Chest tube drainage, blood transfusion requirements, and reexploration rate for nonsurgical bleeding were all increased in the group receiving HES intraoperatively. The authors conclude that HES induces a coagulopathy in cardiac surgery patients, and they no longer use HES intraoperatively in this patient population. A review by Trumble et al. [10] of 85 consecutive subarachnoid hemorrhage patients included 26 patients who developed symptomatic cerebral vasospasm. These patients were treated with hypervolemia, using albumin in 12 patients and HES in 14 patients. The average dose was 1.2 L/day for the duration of symptoms (4-14 days). The PTT in those patients who received HES increased significantly. Although 4 of these 14 patients required transfusion and 2 patients required surgical reexploration for bleeding, compared with a lack of such complications in the albumin group, this was not statistically significant. Finally, the literature contains a handful of case reports describing adverse effects of HES on coagulation [11-15,17,18,42]. In one example, a 13-year-old Jehovah's Witness underwent a posterior spinal fusion. She lost 500 mL of blood and received 1 L of HES intraoperatively and 1 L postoperatively. The patient developed a significant coagulopathy with bleeding from the incision that persisted until the HES was discontinued [12]. In another case, a 49-year-old male who underwent aorto-femoral bypass graft placement received 1 L of HES intraoperatively and three hours postoperatively developed a severe coagulopathy compatible with disseminated intravascular coagulation. He subsequently died of massive hemorrhage [13]. As a third example, a 16-year-old male with epidural and subdural empyema and sepsis was treated with HES. He received approximately 1.5 L of HES in a 24-hour period, shortly thereafter developed disseminated intravascular coagulation, and died [43]. Abramson N. Plasma expanders and bleeding [abstract]. Ann Intern Med 1988;108:307. Mechanism There have been no formal studies on the mechanism by which HES decreases coagulation factor levels or platelet function beyond that observed after hemodilution alone. It is hypothesized that the complex polysaccharide precipitates certain coagulation factors, making them unavailable to the coagulation cascade. HES may form a complex with Factor VIII and fibrinogen, and may also accelerate the action of thrombin in the conversion of fibrinogen to fibrin (the fibrinoplastic effect). This fibrinoplastic effect not only reduces the amount of fibrinogen available for normal coagulation, but also produces a fibrin clot that is lysed more easily. Accordingly, the maximum amplitude of the thromboelastogram was depressed in patients receiving HES, indicating formation of a clot with decreased strength [2]. HES has been suggested to reduce platelet function by coating the platelet surface or inducing platelet damage [1,8,23]. Further research into this matter would be useful. Conclusions Many of the studies cited end with conclusions regarding the effect of HES on coagulation and recommendations for clinical use of the compound. However, as made clear herein, insufficient data are available to formulate formal guidelines regarding the use of HES in specific patients. Many of the studies have design flaws, such as inclusion of nonsurgical or postoperative patients or exclusion of sicker and older patients. These factors, as well as sample size and population studied, may be responsible for lack of demonstrable effect on coagulation or alterations in laboratory values only. Reliable conclusions regarding effects of HES on coagulation and operative bleeding will require an elaborate and large-scale study. Based on data from Boldt et al. [8], a study able to detect a 10% increase in postoperative blood loss in HES-treated, compared with albumin-treated patients, assuming a Type I error of 0.05 and a Type II error of 0.1 (i.e., 90% power), would require more than 200 subjects. Nonetheless, in the absence of such a study, it is reasonable to assume that HES will reduce levels of the coagulation factors fibrinogen, Factor VIII, and von Willebrand's factor, and reduce platelet function. These effects appear to be independent of the dose given. Therefore, it seems difficult to recommend a maximum safe dosage, because patient response may be idiosyncratic, and the manufacturer's guideline of 20 mL/kg is not supported by published data. There are, however, sufficient data to suggest that HES may harm certain patients. Unfortunately, unless a patient has a known coagulopathy (in which case HES should always be avoided; Table 3), it is impossible to know in advance which patients will develop a problem. Even a complete history, physical examination, and assessment of routine laboratory data may not identify such patients. Idiosyncratic responses may occur, surgical technique may vary, and preexisting low levels of a coagulation factor or some other unknown variable can influence outcome. Therefore, although HES is an economically attractive and safe colloid for most patients, there are certain situations in which the clinician should exercise caution when selecting the compound for volume expansion. As always, careful consideration of the patient's pathology and the surgical circumstances are more important than standard rules, especially in view of the limited data available. HES should be used with caution, especially where bleeding would potentially be of serious consequence to the patient (Table 3). Although HES may well be appropriate colloid therapy under these conditions, an additional preoperative evaluation may be necessary before administering the compound (Table 3).Table 3: Caveats for the Use of HESWe hope that additional research on the mechanisms of HES-induced coagulopathy will resolve some of these issues. We gratefully acknowledge Thomas J. Gal, MD, Carl Lynch III, MD, PhD, and Robert M. Epstein, MD, for their helpful comments during preparation of this manuscript. We also acknowledge B. Todd Sitzmann, MD, MPH, for his expert statistical help in calculating required sample sizes for the definitive HES study.