REVIEW Hydroxyethyl starch and granulocyte transfusions: considerations of utility and toxicity profile for patients and donors Daniel R. Ambruso
Hydroxyethyl starch (hetastarch) is a synthetic glucose compound with extensive clinical use as a volume expander. Because of its red blood cell–sedimenting properties, hetastarch plays a major role in preparation of granulocyte products. Recent concerns have been raised about the use of hetastarch in critically ill patients for the development of renal injury and other severe adverse events. In contrast, granulocyte donors receive much less of this compound during collection procedures, and over many years, minimal toxicity has been documented in these individuals. Furthermore, granulocyte products contain very little hetastarch, and ill effects on renal function have not been associated with their administration. This review assesses available information about the toxicity profile of hetastarch in critically ill patients requiring a volume expander as well as granulocyte donors and recipients. Because of the lack of toxicity in these latter two groups, hetastarch should be available for preparation of granulocyte products and their administration.
ABBREVIATIONS: ARF = acute renal failure; ICU = intensive care unit; MS = molar ratio; MW = molecular weight; RRT = renal replacement therapy. From the Departments of Pediatrics and Pathology, University of Colorado Denver, The Anschutz Medical Campus; The Center for Cancer and Blood Disorders and the Transfusion Services, Children’s Hospital Colorado; and the Hematology/Oncology and Bone Marrow Transplantation Laboratories, Aurora, Colorado. Address reprint requests to: Daniel R. Ambruso, MD, East 19th Avenue, Building, Aurora, CO 80245; e-mail: [email protected]. This review was supported by TerumoBCT, Lakewood, CO. The literature review, commentary, and manuscript are the sole work of the author. Received for publication December 4, 2013; revision received August 25, 2014, and accepted August 25, 2014. doi: 10.1111/trf.12892 © 2014 AABB TRANSFUSION 2015;55:911–918. **;**:**-**.
R
ed blood cell (RBC) sedimentation agents have been used since the advent of granulocyte transfusions to remove RBCs and optimize the collection and purification of neutrophils.1 These compounds alter interactions between RBCs allowing aggregation and rouleau formation.2 The most commonly used sedimenting agent, hydroxyethyl starch or hetastarch (HES), has additional characteristics that make it both biologically useful as a nonexpensive colloid for volume replacement and detrimental because of its renal toxicity and potential for inducing bleeding diathesis.3-9 These latter concerns could have practical implications for donors providing granulocyte concentrates or for patients receiving these products. This review will examine the major complications of HES when used in very ill patients as a volume expander and consider issues around the continued use of HES in collection and administration of granulocyte concentrates.
COMPOSITION AND TOXICITY PROFILE HES is a polymeric synthetic glucose compound manufactured by hydrolysis and hydroxyethylation of a highly branched starch, amylopexin. It is characterized by the concentration (usually defined as percent), average molecular weight (MW) of the preparation, the molar ratio (MS) of hydroxyethyl groups to glucose units, and the degree of substitution (ratio of substituted glucose units to total glucose molecules).3-8 The most commonly used concentrations are 3, 6, and 18%; MW, low (70 or 130 kDa), medium (200-270 kDa), and high (450 kDa); and MS low (0.4-0.5) or high (0.6-0.7).8,9 These characteristics relate to the main biologic activity, colloid oncotic pressure; function as a volume expander; and the compound’s retention in the vascular space. In addition, the ratio of hydroxyethylation at C6:C2 may be important to pharmacokinetic characteristics and possibly adverse events.8,9 Whenever possible, preparations will be presented as concentration (%)/MW (kDa)/MS. Pharmacokinetics of HES are complex and, in general, there is a rapid amylase-dependent breakdown and excretion in the urine. When infused, smaller molecules in the less than 60-kDa range are excreted quickly Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION911 1
AMBRUSO
in the urine, and while larger-sized molecules are cleaved by amylase, medium-sized compounds may also be excreted in the bile and feces. The plasma half-life is approximately 5 days and 90% is eliminated in 42 days. A small fraction may be taken up by the reticuloendothelial system, and release from this site occurs slowly. Small amounts of various preparations can be detected for several weeks. The survival of different HES preparations is between 96 hours for medium MW and 17 to 26 weeks or longer for high-MW compounds.10-12 Adverse events reported, primarily for use of HES as a volume expander, include hypersensitivity reactions (urticaria and more generalized symptoms and signs of anaphylaxis); circulatory overload and its complications; and coagulopathy, pruritus, and other minor effects.3-8 HES-induced coagulopathy involves multiple aspects of hemostasis.13 A major alteration is attributed to dilutional effects. In addition, these colloid solutions may impair platelet (PLT) function by a variety of mechanisms including blocking access of surface receptors to their ligands and interference with intracellular signal transduction. von Willebrand factor and Factor VIII levels may also be adversely affected, fibrin polymerization decreased, and hypercoagulability induced by HES. These effects are in addition to the hemostatic complications of the patients’ underlying disease state and clinical status. Hypersensitivity reactions are uncommon (0.01%-0.1%). Other reactions such as headache and volume overload occur in not more than 10% of patients and may be dose related. Lower MW preparations of HES exhibit fewer complications than the high-MW HES, particularly those related to coagulopathy and PLT dysfunction.14 HES has a low-risk adverse event profile for granulocyte donors and recipients. The association of HES with mortality and renal toxicities has been described in critically ill patients and is discussed in detail below.
REPORTS OF RENAL TOXICITY IN CRITICALLY ILL PATIENTS HES use in critically ill patients is part of a longstanding crystalloid/colloid debate concerning the most appropriate volume resuscitation fluid. HES was developed as a replacement for 5% albumin solution.9,15 Within this context, HES has become widely used because it is inexpensive and has a clinical role as a volume expander.9 Although the development of renal dysfunction associated with its use is not included in its toxicology profile, reports from the literature have demonstrated that patients infused with HES may experience altered renal function and other adverse events. Our understanding of populations at risk for these effects is incomplete. The mechanism by which HES may cause renal dysfunction is not well understood. The lower-MW (50-kDa) HES preparations are cleared directly by glomerular 912 TRANSFUSION Volume Volume April 2015 2 TRANSFUSION **,55, ** **
filtration, but larger molecules may be hydrolyzed by α-amylase before clearance by this mechanism is affected.3-9 Toxicology studies have suggested that reabsorption of HES by renal tubular cells may cause reversible swelling leading to tubular obstruction, both risk factors for acute renal failure (ARF).9,16 In addition, the presence of hyperoncotic molecules could result in hyperviscosity and stasis of tubular flow.17 These latter effects could be ascribed to any colloid solution used in patients who may be dehydrated or severely volume depleted and require fluid loading.18 Earlier retrospective studies evaluated several patient groups to explore renal toxicity of HES given as a volume expander. Three studies examining the effects of HES infusion in brain-dead kidney donors demonstrated a range of complications from nephrosis-like lesions and mild increases in creatinine to increased incidence of hemodialysis in transplanted kidneys.19-21 Additional studies explored the use of HES perioperatively in patients without prior renal dysfunction.22-24 Two studies of cardiothoracic surgery using medium- and low-MW HES noted that no change in renal function was noted.23,24 Patients undergoing middle ear surgery receiving different HES preparations exhibited no renal dysfunction measured by sensitive marker studies.25 In a randomized, prospective study of patients undergoing major abdominal surgery receiving 1300 to 3000 mL, low-MW HES or medium-MW HES exhibited no abnormalities.26 Several studies evaluated the use of HES in intensive care unit (ICU) patient populations including patients requiring volume expansion after cardiac surgery, individuals with sepsis or septic shock and trauma, and postoperative patients with sepsis.23,27-29 Of these studies, one comparing HES with albumin in patients with sepsis or septic shock at three centers showed a significant increase in ARF, oliguria, and peak creatinine in the high-MW-HES– treated group.27 The remaining studies claimed HES preparations were as effective as albumin or gelatin and without significant renal toxicity.23,28,29 No specific information about volumes required for HES to induce ARF is available. Data from studies noted above suggest that, in adults, volumes of 1 to 3 L may be associated with specific renal effects.20,29 However, in a single case, a dose of less than 10 mL/kg is suspected of causing ARF,30 and conversely, administration of as much as 3 L of HES does not alter renal function.29 Toxicity in patients with small doses, not more than 10 mL/kg, represents the exception rather than the rule. Furthermore, it is not clear what effect repetitive doses over sequential days would have on renal function. In the past few years there have been additional studies addressing safety concerns about the use of HES in critically ill patients. Thirteen studies focus on risks of kidney injury or dysfunction, mortality, bleeding, and other events in a variety of different patient groups. These
TOXICITY TOXICITY PROFILE PROFILE OF OF HES HES FOR FOR GRANULOCYTE GRANULOCYTE TRANSFUSIONS TRANSFUSIONS
can be separated into two categories, observational, retrospective, or controlled trials (n = 7) and reviews of multiple studies with meta-analyses (n = 6). A summary of the findings of these studies is shown in Table 1.31-43 Included is a description of the study characteristics and results as they relate to renal status, mortality, and other adverse events. Details may be found in the table or in the specific references. Of the observational, retrospective, or controlled trials,31-37 the investigations analyze subjects with sepsis (three), cardiac surgery (one), trauma (one), and ICU admissions (one) and after resuscitation (one). Six of the seven studies documented significant increase in at least one renal complication with administration of HES, kidney injury or risk for injury, increase in creatinine, development of ARF, requirement for renal replacement therapy (RRT), and risk for kidney injury.32-37 The most frequent were the requirement for RRT (three of seven) and evidence of increased kidney injury (four of seven). Two specifically examined ARF and documented no increase with HES treatment.34,36 One article claimed that HES had no influence on renal function.31 Two evaluations demonstrated increased mortality33,35 and two observed no increase.34,36 Two articles claimed an increase in other adverse events, including bleeding, rash, and pruritus,34,35 while one claimed no increase in these adverse events.36 One study stated that there was no coagulopathy in patients receiving HES36 while another described more blood products used.37 The meta-analyses38-43 included studies of cardiothoracic surgery (one), ill patients requiring resuscitation (one), critically ill individuals (two), and patients with sepsis (two). In five of six summaries, renal complications were increased in HES-treated subjects38,40-43 with all five describing increased requirement for RRT.38,40-43 Two studies showed an increase in ARF38,41 and one with an increased risk for renal injury.40 A single study claimed that HES-treated patients had more other serious adverse events40 and two stated that more patients were treated with RBCs40 or had increased blood loss or rebleeding or needed more blood products.39 Finally, two articles showed increased mortality41,43 with HES treatment while three demonstrated no difference in comparison to control.39,40,42 While information from meta-analyses summarizes results from large numbers of patients, differences in study groups, details, approaches, and study design may limit the comparability and the specific information obtained as well as the conclusions reached from the analyses. However, from all the clinical information provided by the studies above, in critically ill ICU patients and those with surgery, trauma, and sepsis, use of HES as a volume expander may be associated with mortality, increased renal complications, and other adverse events such as coagulopathy and bleeding. It appears that renal
dysfunction associated with HES administration is likely to occur in severely ill patients whose primary clinical condition predisposes to renal failure including hypovolemia, shock, sepsis, and dehydration. There is no direct dose–response relationship between HES and renal dysfunction, but adult patients receiving 1 to 3 L acutely may be more likely to require RRT or suffer from renal dysfunction or ARF. Because of the data presented in the most recent studies, actions have been taken to limit the use of HES as a volume expander in critically ill patients including those with sepsis and admission to the ICU with other critical conditions. On June 14, 2013, EU drug regulators on the European Medicines Agencies Pharmacovigilance Risk Assessment Committee (PRAC) recommended pulling HES solutions from various national markets because evidence shows that the benefits of these preparations no longer outweigh their risks (http://www.ema.europa.eu/ ema/index.jsp?curl=pages/news_and_events/20). This recommendation will be presented to the Coordination Group for Mutual Recognition and Decentralization Procedures-Human for final decision. On June 24, 2013, after reviewing the data described above, the FDA concluded that HES solutions should not be used in critically ill patients, including those with sepsis and those admitted to the ICU, and a boxed warning to include the risk of mortality and severe renal injury was warranted. In addition, the FDA has reviewed the metaanalysis bleeding data for patients undergoing open heart surgery in association with cardiopulmonary bypass and has determined that additional comments about excessive bleeding are needed in the warnings and precautions section of the package insert. The following is a summary of their recommendations: http://www.fda.gov/Biologics BloodVaccines/SafetyAvailability/ucm358271.htm
Recommendations for health professionals •
• • •
•
•
Do not use HES solutions in critically ill adult patients including those with sepsis, and those admitted to the ICU. Avoid use in patients with existing renal dysfunction. Discontinue use of HES at the first sign of renal injury. Need for RRT has been reported up to 90 days after HES administration. Continue to monitor renal function for at least 90 days in all patients. Avoid use in patients undergoing open heart surgery in association with cardiopulmonary bypass due to excess bleeding. Discontinue use of HES at the first sign of coagulopathy.
Additional recommendations are made for patients to be made aware of the risks for HES usage and discuss them with their healthcare provider. Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION913 3
AMBRUSO
TABLE 1. Recent studies of the safety of HES as a volume expander in critically ill patients Author, year, reference
Study design Observational study. 198 ICUs enrolled 3,147 patients. All admitted patients studied.
Treatment groups 1075 patients received HES. These patients more likely to have oncology problems, CHF, higher illness rating scores.
Rioux, 200932
Retrospective review of cardiac surgery patients.
563 patients received pentastarch (10%; 250/0.45). No controls.
Lissauer, 201133
Retrospective review. Adult trauma patients, 2225 subjects.
497 received 6% HES (450/0.7) in first 24 hours. Remainder did not receive HES.
Myburgh, 201234
Randomized trial. Half patients received 6% HES, 130/0.4 (Volvulen) vs. normal saline for fluid resuscitation.
7000 patients. Primary outcome, death at 90 days; secondary outcomes, acute kidney injury/failure and RRT. 3315 in HES-treated group and 3336 in the saline group.
Perner, 201235
Multicenter, parallel-group, blinded trial. Primary outcomes death or renal failure at 90 days. All patients with severe sepsis.
804 patients treated, 798 intention to treat. Fluid resuscitation in ICU, 6% HES 130/0.42 (tetraspan) vs. lactated Ringer’s, 33 mL/kg. 398 patients assigned to HES and 400 patients to Ringer’s lactate.
Guidet, 201236
Multicenter, prospective, active-controlled, double-blind, randomized study for patients with sepsis.
6% HES 130/0.41 vs. 0.9% saline used to achieve hemodynamic stability (HDS). 196 patients studied; 174 reached HDS.
Bayer, 201237
Prospective, single-center study of consecutive patients with sepsis in the surgical ICU.
Evaluated impaired renal function in patients treated with 6% HES 130/0.4 (360), 4% gelatin (352), or crystalloid (334).
Dart, 201038
Review of RCTs or quasi-RCTs of fluids to prevent volume depletion. Goal was to examine effects of HES on kidney function.
34 studies involving 2607 patients. Comparison of HES vs. other fluids. Evaluated RRT, acute kidney injury, creatinine, and creatinine clearance.
Sakr, 200731
914 TRANSFUSION Volume Volume April 2015 4 TRANSFUSION **,55, ** **
Results More RRT required in HES than non-HES patients (p < 0.006). Administration was not associated with increased risk for subsequent RRT. HES had no influence in renal function. 10% of patients developed acute kidney injury defined by 50% increase in creatinine. Pentastarch infusion was independent predictor of injury (p < 0.001). Those who developed injury received more pentastarch, 16 ± 9 mL/kg vs. 10 ± 7 mL/kg (p < 0.001). Pentastarch a risk factor for renal injury until first day postoperatively. HES associated with acute kidney injury (13% with HES vs. 8% without, RR 1.73; 95% CI, 1.30-2.28). HES also associated with renal dysfunction. HES associated with increased mortality, 21% vs. 11% (RR, 1.84; 95% CI, 1.48-2.29). Authors suggest avoiding resuscitation with HES in trauma. No difference in 90-day mortality. RRT used in 7% of HES group vs. 5.8% in saline group (p = 0.04). Renal injury greater in saline group (p = 0.005). Renal failure not different. HES associated with more adverse events (p < 0.001) including rash and pruritus. 51% HES-treated patients died vs. 43% with lactated Ringer’s (p = 0.03). 22% with HES had RRT vs. 16% control (p = 0.04). 10% with HES had severe bleeding vs. 6% control (p = 0.09). HES increased risk of death and RRT. Less HES used vs. saline (p = 0.0185). Time to HDS shorter with HES. ARF in 24.5% HES treated vs. 20% saline (NS). No difference in renal dysfunction scores, mortality, coagulopathy, or pruritus. Hospital stay, severity scores no different between groups. Similar time to shock reversal. HES was an independent risk factor for acute kidney injury (OR [95% CI], 2.55 [1.76-3.69]) and received more blood products. Acute kidney failure and need for RRT higher in HES-treated patients, RR 1.5 (95% CI, 1.2-1.7) and 1.38 (0.89-2.16) respectively compared to other fluid therapies.
TOXICITY TOXICITY PROFILE PROFILE OF OF HES HES FOR FOR GRANULOCYTE GRANULOCYTE TRANSFUSIONS TRANSFUSIONS
TABLE 1. Continued Author, year, reference
Study design Meta-analysis of randomized clinical trials. Impact of HES vs. albumin on postoperative blood loss in adult, cardiopulmonary bypass surgery.
Treatment groups HES-treated patients vs. albumin. 18 trials analyzed 970 patients.
Haase, 201340
Systematic review with meta-analysis and sequential analysis of RCTs. Evaluated patients with sepsis.
3456 patients with sepsis in nine trials. Comparison of treatment with HES 130/0.38-0.45 vs. crystalloid or albumin.
Zarychanski, 201341
Systematic review of RCTs; excluded seven studies that had been retracted.
Critically ill patients requiring volume resuscitation. 10,290 patients evaluated, excluding patients from retracted studies. 31 trials.
Gattas, 201342
Systematic review and meta-analysis of RCTs on mortality and RRT. Evaluated resuscitation fluids in critically ill adults.
35 trials evaluated. 10,391 participants. Comparison of 6% HES 130/0.4 vs. other fluids.
Patel, 201343
Systematic review and meta-analysis of RRTs in severe sepsis. Primary outcome, 90-day mortality.
6 RCTs. 3033 patients. Comparison 6% HES 130/0.4 with crystalloid fluid resuscitation. Median dose 37.4 mL/kg.
Navickis, 201239
Results HES increased postoperative blood loss by 33% (p < 0.001). Risk of reoperation for bleeding doubled (p = 0.02). Increased transfusion of RBCs by 28% (p < 0.001), FFP by 30% (p < 0.008), PLTs by 29% (p = 0.027). No differences in mortality. No increase in risk of death between the groups. RRT used more with HES treatment (RR, 1.36; 95% CI, 1.03-1.80). RR of renal injury increased (1.18; 95% CI, 0.99-1.40). More patients with HES treated with RBCs and sustained serious adverse events. Volume of RBCs no different between groups. HES associated with increased mortality (RR, 1.09; 95% CI, 1.02-1.17). There was an increase in renal failure (RR, 1.27; 95% CI, 1.09-147) and requirement for RRT (RR, 1.32; 95% CI, 1.15-1.50). HES use in critically ill patients not warranted. Death in 19.8% HES-treated patients vs. 18.5% controls (NS). Treatment with RRT occurred in 8.9% HES-treated vs. 7.2% of others (RR, 1.25; 95% CI, 1.08-1.44). Patients treated with HES have an increase in RRT. 90-day mortality increased in HES group (p = 0.02). HES associated with increased RRT (p = 0.001). 28-day mortality not different but overall mortality was different (p = 0.02). HES should be avoided for resuscitation in severe sepsis.
CHF = congestive heart failure; RCT = randomized controlled trial; RR = relative risk.
IMPLICATIONS OF HES IN HEALTHY GRANULOCYTE DONORS For many years macromolecular compounds have been used in collecting granulocyte products to aid in sedimentation of RBCs away from neutrophils to promote purity of the concentrates and enhance the white blood cell (WBC) yield.1,10 Although in the early days of granulocyte collections dextran, gelatin, or HES was used, most centers engaged in collections switched to two main preparations, 6% HES/450/0.7 (hetastarch) or 10% HES/264/0.45 (pentastarch).1,10 The optimal product used was controversial, and early studies claimed comparable efficiencies and granulocyte yields. Kinetic variables and clearance from individuals as well as the toxicity profiles favored pentastarch.44 However, the efficacy of pentastarch for
granulocyte collection was challenged in two clinical studies.11,45 In the former trial in which steroid-treated donors underwent paired granulocyte collections with 6% HES/450/0.7 or 10% HES/264/0.45, in a standardized protocol processing 7 L with 500 mL of the HES preparation, the mean yield was much greater with the former sedimenting agent.11 The enhanced yields by hetastarch were confirmed in several recent studies of granulocyte– colony-stimulating factor (G-CSF)–mobilized donors.46-49 The negative implications for HES use in granulocyte donors are minimal. In general, donors undergo procedures at monthly intervals but this may be more frequent with 1 to 2 weeks as intervals between collections. With each procedure, approximately 7 L of donor blood volume is processed1 with a range in published studies of 7 to 10 L, and 500 mL but usually no more than 750 mL of HES is Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION915 5
AMBRUSO
used in the collection. The actual amount the donors receive is approximately 450 to 475 mL as some of the HES remains with the collected granulocytes. This is less than the usual dose given to critically ill patients as a volume expander in the previously reviewed studies. Even if multiple collections were performed with up to two or three donations in a week, the risk for renal problems is very low in the donor population After single apheresis procedures, levels of HES decrease rapidly with a half-life up to 3.8 days and a longer time (6-7 weeks) for complete elimination.11,12 With multiple procedures, the pattern is similar, but persistence of trace amounts of HES may occur up to 72 weeks. Although long-term effects are not precisely known, it seems less likely they will occur on the basis of dose administered and excretion the majority of the drug. In studies of granulocyte donors, the side effects reported with administration of HES, separate from any adverse events related to mobilization with steroids and/or G-CSF, are minimal.11,44-52 Most series have documented pruritus as a complication in a low number of donors (up to 6% in one series, not described in others) and collections (0.1%-0.7%). Consequences related to HES effects on volume or renal dysfunction have not been documented, and no incidents of mortality have been described. A postmarketing review of COBE Spectra system complaints over 10 years from 2001 to 2010 covering approximately 40,000 granulocyte collections did not reveal any serious adverse donor events or deaths (TerumoBCT Spectra Optia apheresis system, personal communications from a clinical evaluation report, 2012). This included serious events related to HES used in the collections. The search revealed only minor device malfunctions. Other adverse reactions (e.g., anaphylaxis, bleeding complications, or others) are either very rare (
Hydroxyethyl starch (hetastarch) is a synthetic glucose compound with extensive clinical use as a volume expander. Because of its red blood cell–sedimenting properties, hetastarch plays a major role in preparation of granulocyte products. Recent concerns have been raised about the use of hetastarch in critically ill patients for the development of renal injury and other severe adverse events. In contrast, granulocyte donors receive much less of this compound during collection procedures, and over many years, minimal toxicity has been documented in these individuals. Furthermore, granulocyte products contain very little hetastarch, and ill effects on renal function have not been associated with their administration. This review assesses available information about the toxicity profile of hetastarch in critically ill patients requiring a volume expander as well as granulocyte donors and recipients. Because of the lack of toxicity in these latter two groups, hetastarch should be available for preparation of granulocyte products and their administration.
ABBREVIATIONS: ARF = acute renal failure; ICU = intensive care unit; MS = molar ratio; MW = molecular weight; RRT = renal replacement therapy. From the Departments of Pediatrics and Pathology, University of Colorado Denver, The Anschutz Medical Campus; The Center for Cancer and Blood Disorders and the Transfusion Services, Children’s Hospital Colorado; and the Hematology/Oncology and Bone Marrow Transplantation Laboratories, Aurora, Colorado. Address reprint requests to: Daniel R. Ambruso, MD, East 19th Avenue, Building, Aurora, CO 80245; e-mail: [email protected]. This review was supported by TerumoBCT, Lakewood, CO. The literature review, commentary, and manuscript are the sole work of the author. Received for publication December 4, 2013; revision received August 25, 2014, and accepted August 25, 2014. doi: 10.1111/trf.12892 © 2014 AABB TRANSFUSION 2015;55:911–918. **;**:**-**.
R
ed blood cell (RBC) sedimentation agents have been used since the advent of granulocyte transfusions to remove RBCs and optimize the collection and purification of neutrophils.1 These compounds alter interactions between RBCs allowing aggregation and rouleau formation.2 The most commonly used sedimenting agent, hydroxyethyl starch or hetastarch (HES), has additional characteristics that make it both biologically useful as a nonexpensive colloid for volume replacement and detrimental because of its renal toxicity and potential for inducing bleeding diathesis.3-9 These latter concerns could have practical implications for donors providing granulocyte concentrates or for patients receiving these products. This review will examine the major complications of HES when used in very ill patients as a volume expander and consider issues around the continued use of HES in collection and administration of granulocyte concentrates.
COMPOSITION AND TOXICITY PROFILE HES is a polymeric synthetic glucose compound manufactured by hydrolysis and hydroxyethylation of a highly branched starch, amylopexin. It is characterized by the concentration (usually defined as percent), average molecular weight (MW) of the preparation, the molar ratio (MS) of hydroxyethyl groups to glucose units, and the degree of substitution (ratio of substituted glucose units to total glucose molecules).3-8 The most commonly used concentrations are 3, 6, and 18%; MW, low (70 or 130 kDa), medium (200-270 kDa), and high (450 kDa); and MS low (0.4-0.5) or high (0.6-0.7).8,9 These characteristics relate to the main biologic activity, colloid oncotic pressure; function as a volume expander; and the compound’s retention in the vascular space. In addition, the ratio of hydroxyethylation at C6:C2 may be important to pharmacokinetic characteristics and possibly adverse events.8,9 Whenever possible, preparations will be presented as concentration (%)/MW (kDa)/MS. Pharmacokinetics of HES are complex and, in general, there is a rapid amylase-dependent breakdown and excretion in the urine. When infused, smaller molecules in the less than 60-kDa range are excreted quickly Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION911 1
AMBRUSO
in the urine, and while larger-sized molecules are cleaved by amylase, medium-sized compounds may also be excreted in the bile and feces. The plasma half-life is approximately 5 days and 90% is eliminated in 42 days. A small fraction may be taken up by the reticuloendothelial system, and release from this site occurs slowly. Small amounts of various preparations can be detected for several weeks. The survival of different HES preparations is between 96 hours for medium MW and 17 to 26 weeks or longer for high-MW compounds.10-12 Adverse events reported, primarily for use of HES as a volume expander, include hypersensitivity reactions (urticaria and more generalized symptoms and signs of anaphylaxis); circulatory overload and its complications; and coagulopathy, pruritus, and other minor effects.3-8 HES-induced coagulopathy involves multiple aspects of hemostasis.13 A major alteration is attributed to dilutional effects. In addition, these colloid solutions may impair platelet (PLT) function by a variety of mechanisms including blocking access of surface receptors to their ligands and interference with intracellular signal transduction. von Willebrand factor and Factor VIII levels may also be adversely affected, fibrin polymerization decreased, and hypercoagulability induced by HES. These effects are in addition to the hemostatic complications of the patients’ underlying disease state and clinical status. Hypersensitivity reactions are uncommon (0.01%-0.1%). Other reactions such as headache and volume overload occur in not more than 10% of patients and may be dose related. Lower MW preparations of HES exhibit fewer complications than the high-MW HES, particularly those related to coagulopathy and PLT dysfunction.14 HES has a low-risk adverse event profile for granulocyte donors and recipients. The association of HES with mortality and renal toxicities has been described in critically ill patients and is discussed in detail below.
REPORTS OF RENAL TOXICITY IN CRITICALLY ILL PATIENTS HES use in critically ill patients is part of a longstanding crystalloid/colloid debate concerning the most appropriate volume resuscitation fluid. HES was developed as a replacement for 5% albumin solution.9,15 Within this context, HES has become widely used because it is inexpensive and has a clinical role as a volume expander.9 Although the development of renal dysfunction associated with its use is not included in its toxicology profile, reports from the literature have demonstrated that patients infused with HES may experience altered renal function and other adverse events. Our understanding of populations at risk for these effects is incomplete. The mechanism by which HES may cause renal dysfunction is not well understood. The lower-MW (50-kDa) HES preparations are cleared directly by glomerular 912 TRANSFUSION Volume Volume April 2015 2 TRANSFUSION **,55, ** **
filtration, but larger molecules may be hydrolyzed by α-amylase before clearance by this mechanism is affected.3-9 Toxicology studies have suggested that reabsorption of HES by renal tubular cells may cause reversible swelling leading to tubular obstruction, both risk factors for acute renal failure (ARF).9,16 In addition, the presence of hyperoncotic molecules could result in hyperviscosity and stasis of tubular flow.17 These latter effects could be ascribed to any colloid solution used in patients who may be dehydrated or severely volume depleted and require fluid loading.18 Earlier retrospective studies evaluated several patient groups to explore renal toxicity of HES given as a volume expander. Three studies examining the effects of HES infusion in brain-dead kidney donors demonstrated a range of complications from nephrosis-like lesions and mild increases in creatinine to increased incidence of hemodialysis in transplanted kidneys.19-21 Additional studies explored the use of HES perioperatively in patients without prior renal dysfunction.22-24 Two studies of cardiothoracic surgery using medium- and low-MW HES noted that no change in renal function was noted.23,24 Patients undergoing middle ear surgery receiving different HES preparations exhibited no renal dysfunction measured by sensitive marker studies.25 In a randomized, prospective study of patients undergoing major abdominal surgery receiving 1300 to 3000 mL, low-MW HES or medium-MW HES exhibited no abnormalities.26 Several studies evaluated the use of HES in intensive care unit (ICU) patient populations including patients requiring volume expansion after cardiac surgery, individuals with sepsis or septic shock and trauma, and postoperative patients with sepsis.23,27-29 Of these studies, one comparing HES with albumin in patients with sepsis or septic shock at three centers showed a significant increase in ARF, oliguria, and peak creatinine in the high-MW-HES– treated group.27 The remaining studies claimed HES preparations were as effective as albumin or gelatin and without significant renal toxicity.23,28,29 No specific information about volumes required for HES to induce ARF is available. Data from studies noted above suggest that, in adults, volumes of 1 to 3 L may be associated with specific renal effects.20,29 However, in a single case, a dose of less than 10 mL/kg is suspected of causing ARF,30 and conversely, administration of as much as 3 L of HES does not alter renal function.29 Toxicity in patients with small doses, not more than 10 mL/kg, represents the exception rather than the rule. Furthermore, it is not clear what effect repetitive doses over sequential days would have on renal function. In the past few years there have been additional studies addressing safety concerns about the use of HES in critically ill patients. Thirteen studies focus on risks of kidney injury or dysfunction, mortality, bleeding, and other events in a variety of different patient groups. These
TOXICITY TOXICITY PROFILE PROFILE OF OF HES HES FOR FOR GRANULOCYTE GRANULOCYTE TRANSFUSIONS TRANSFUSIONS
can be separated into two categories, observational, retrospective, or controlled trials (n = 7) and reviews of multiple studies with meta-analyses (n = 6). A summary of the findings of these studies is shown in Table 1.31-43 Included is a description of the study characteristics and results as they relate to renal status, mortality, and other adverse events. Details may be found in the table or in the specific references. Of the observational, retrospective, or controlled trials,31-37 the investigations analyze subjects with sepsis (three), cardiac surgery (one), trauma (one), and ICU admissions (one) and after resuscitation (one). Six of the seven studies documented significant increase in at least one renal complication with administration of HES, kidney injury or risk for injury, increase in creatinine, development of ARF, requirement for renal replacement therapy (RRT), and risk for kidney injury.32-37 The most frequent were the requirement for RRT (three of seven) and evidence of increased kidney injury (four of seven). Two specifically examined ARF and documented no increase with HES treatment.34,36 One article claimed that HES had no influence on renal function.31 Two evaluations demonstrated increased mortality33,35 and two observed no increase.34,36 Two articles claimed an increase in other adverse events, including bleeding, rash, and pruritus,34,35 while one claimed no increase in these adverse events.36 One study stated that there was no coagulopathy in patients receiving HES36 while another described more blood products used.37 The meta-analyses38-43 included studies of cardiothoracic surgery (one), ill patients requiring resuscitation (one), critically ill individuals (two), and patients with sepsis (two). In five of six summaries, renal complications were increased in HES-treated subjects38,40-43 with all five describing increased requirement for RRT.38,40-43 Two studies showed an increase in ARF38,41 and one with an increased risk for renal injury.40 A single study claimed that HES-treated patients had more other serious adverse events40 and two stated that more patients were treated with RBCs40 or had increased blood loss or rebleeding or needed more blood products.39 Finally, two articles showed increased mortality41,43 with HES treatment while three demonstrated no difference in comparison to control.39,40,42 While information from meta-analyses summarizes results from large numbers of patients, differences in study groups, details, approaches, and study design may limit the comparability and the specific information obtained as well as the conclusions reached from the analyses. However, from all the clinical information provided by the studies above, in critically ill ICU patients and those with surgery, trauma, and sepsis, use of HES as a volume expander may be associated with mortality, increased renal complications, and other adverse events such as coagulopathy and bleeding. It appears that renal
dysfunction associated with HES administration is likely to occur in severely ill patients whose primary clinical condition predisposes to renal failure including hypovolemia, shock, sepsis, and dehydration. There is no direct dose–response relationship between HES and renal dysfunction, but adult patients receiving 1 to 3 L acutely may be more likely to require RRT or suffer from renal dysfunction or ARF. Because of the data presented in the most recent studies, actions have been taken to limit the use of HES as a volume expander in critically ill patients including those with sepsis and admission to the ICU with other critical conditions. On June 14, 2013, EU drug regulators on the European Medicines Agencies Pharmacovigilance Risk Assessment Committee (PRAC) recommended pulling HES solutions from various national markets because evidence shows that the benefits of these preparations no longer outweigh their risks (http://www.ema.europa.eu/ ema/index.jsp?curl=pages/news_and_events/20). This recommendation will be presented to the Coordination Group for Mutual Recognition and Decentralization Procedures-Human for final decision. On June 24, 2013, after reviewing the data described above, the FDA concluded that HES solutions should not be used in critically ill patients, including those with sepsis and those admitted to the ICU, and a boxed warning to include the risk of mortality and severe renal injury was warranted. In addition, the FDA has reviewed the metaanalysis bleeding data for patients undergoing open heart surgery in association with cardiopulmonary bypass and has determined that additional comments about excessive bleeding are needed in the warnings and precautions section of the package insert. The following is a summary of their recommendations: http://www.fda.gov/Biologics BloodVaccines/SafetyAvailability/ucm358271.htm
Recommendations for health professionals •
• • •
•
•
Do not use HES solutions in critically ill adult patients including those with sepsis, and those admitted to the ICU. Avoid use in patients with existing renal dysfunction. Discontinue use of HES at the first sign of renal injury. Need for RRT has been reported up to 90 days after HES administration. Continue to monitor renal function for at least 90 days in all patients. Avoid use in patients undergoing open heart surgery in association with cardiopulmonary bypass due to excess bleeding. Discontinue use of HES at the first sign of coagulopathy.
Additional recommendations are made for patients to be made aware of the risks for HES usage and discuss them with their healthcare provider. Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION913 3
AMBRUSO
TABLE 1. Recent studies of the safety of HES as a volume expander in critically ill patients Author, year, reference
Study design Observational study. 198 ICUs enrolled 3,147 patients. All admitted patients studied.
Treatment groups 1075 patients received HES. These patients more likely to have oncology problems, CHF, higher illness rating scores.
Rioux, 200932
Retrospective review of cardiac surgery patients.
563 patients received pentastarch (10%; 250/0.45). No controls.
Lissauer, 201133
Retrospective review. Adult trauma patients, 2225 subjects.
497 received 6% HES (450/0.7) in first 24 hours. Remainder did not receive HES.
Myburgh, 201234
Randomized trial. Half patients received 6% HES, 130/0.4 (Volvulen) vs. normal saline for fluid resuscitation.
7000 patients. Primary outcome, death at 90 days; secondary outcomes, acute kidney injury/failure and RRT. 3315 in HES-treated group and 3336 in the saline group.
Perner, 201235
Multicenter, parallel-group, blinded trial. Primary outcomes death or renal failure at 90 days. All patients with severe sepsis.
804 patients treated, 798 intention to treat. Fluid resuscitation in ICU, 6% HES 130/0.42 (tetraspan) vs. lactated Ringer’s, 33 mL/kg. 398 patients assigned to HES and 400 patients to Ringer’s lactate.
Guidet, 201236
Multicenter, prospective, active-controlled, double-blind, randomized study for patients with sepsis.
6% HES 130/0.41 vs. 0.9% saline used to achieve hemodynamic stability (HDS). 196 patients studied; 174 reached HDS.
Bayer, 201237
Prospective, single-center study of consecutive patients with sepsis in the surgical ICU.
Evaluated impaired renal function in patients treated with 6% HES 130/0.4 (360), 4% gelatin (352), or crystalloid (334).
Dart, 201038
Review of RCTs or quasi-RCTs of fluids to prevent volume depletion. Goal was to examine effects of HES on kidney function.
34 studies involving 2607 patients. Comparison of HES vs. other fluids. Evaluated RRT, acute kidney injury, creatinine, and creatinine clearance.
Sakr, 200731
914 TRANSFUSION Volume Volume April 2015 4 TRANSFUSION **,55, ** **
Results More RRT required in HES than non-HES patients (p < 0.006). Administration was not associated with increased risk for subsequent RRT. HES had no influence in renal function. 10% of patients developed acute kidney injury defined by 50% increase in creatinine. Pentastarch infusion was independent predictor of injury (p < 0.001). Those who developed injury received more pentastarch, 16 ± 9 mL/kg vs. 10 ± 7 mL/kg (p < 0.001). Pentastarch a risk factor for renal injury until first day postoperatively. HES associated with acute kidney injury (13% with HES vs. 8% without, RR 1.73; 95% CI, 1.30-2.28). HES also associated with renal dysfunction. HES associated with increased mortality, 21% vs. 11% (RR, 1.84; 95% CI, 1.48-2.29). Authors suggest avoiding resuscitation with HES in trauma. No difference in 90-day mortality. RRT used in 7% of HES group vs. 5.8% in saline group (p = 0.04). Renal injury greater in saline group (p = 0.005). Renal failure not different. HES associated with more adverse events (p < 0.001) including rash and pruritus. 51% HES-treated patients died vs. 43% with lactated Ringer’s (p = 0.03). 22% with HES had RRT vs. 16% control (p = 0.04). 10% with HES had severe bleeding vs. 6% control (p = 0.09). HES increased risk of death and RRT. Less HES used vs. saline (p = 0.0185). Time to HDS shorter with HES. ARF in 24.5% HES treated vs. 20% saline (NS). No difference in renal dysfunction scores, mortality, coagulopathy, or pruritus. Hospital stay, severity scores no different between groups. Similar time to shock reversal. HES was an independent risk factor for acute kidney injury (OR [95% CI], 2.55 [1.76-3.69]) and received more blood products. Acute kidney failure and need for RRT higher in HES-treated patients, RR 1.5 (95% CI, 1.2-1.7) and 1.38 (0.89-2.16) respectively compared to other fluid therapies.
TOXICITY TOXICITY PROFILE PROFILE OF OF HES HES FOR FOR GRANULOCYTE GRANULOCYTE TRANSFUSIONS TRANSFUSIONS
TABLE 1. Continued Author, year, reference
Study design Meta-analysis of randomized clinical trials. Impact of HES vs. albumin on postoperative blood loss in adult, cardiopulmonary bypass surgery.
Treatment groups HES-treated patients vs. albumin. 18 trials analyzed 970 patients.
Haase, 201340
Systematic review with meta-analysis and sequential analysis of RCTs. Evaluated patients with sepsis.
3456 patients with sepsis in nine trials. Comparison of treatment with HES 130/0.38-0.45 vs. crystalloid or albumin.
Zarychanski, 201341
Systematic review of RCTs; excluded seven studies that had been retracted.
Critically ill patients requiring volume resuscitation. 10,290 patients evaluated, excluding patients from retracted studies. 31 trials.
Gattas, 201342
Systematic review and meta-analysis of RCTs on mortality and RRT. Evaluated resuscitation fluids in critically ill adults.
35 trials evaluated. 10,391 participants. Comparison of 6% HES 130/0.4 vs. other fluids.
Patel, 201343
Systematic review and meta-analysis of RRTs in severe sepsis. Primary outcome, 90-day mortality.
6 RCTs. 3033 patients. Comparison 6% HES 130/0.4 with crystalloid fluid resuscitation. Median dose 37.4 mL/kg.
Navickis, 201239
Results HES increased postoperative blood loss by 33% (p < 0.001). Risk of reoperation for bleeding doubled (p = 0.02). Increased transfusion of RBCs by 28% (p < 0.001), FFP by 30% (p < 0.008), PLTs by 29% (p = 0.027). No differences in mortality. No increase in risk of death between the groups. RRT used more with HES treatment (RR, 1.36; 95% CI, 1.03-1.80). RR of renal injury increased (1.18; 95% CI, 0.99-1.40). More patients with HES treated with RBCs and sustained serious adverse events. Volume of RBCs no different between groups. HES associated with increased mortality (RR, 1.09; 95% CI, 1.02-1.17). There was an increase in renal failure (RR, 1.27; 95% CI, 1.09-147) and requirement for RRT (RR, 1.32; 95% CI, 1.15-1.50). HES use in critically ill patients not warranted. Death in 19.8% HES-treated patients vs. 18.5% controls (NS). Treatment with RRT occurred in 8.9% HES-treated vs. 7.2% of others (RR, 1.25; 95% CI, 1.08-1.44). Patients treated with HES have an increase in RRT. 90-day mortality increased in HES group (p = 0.02). HES associated with increased RRT (p = 0.001). 28-day mortality not different but overall mortality was different (p = 0.02). HES should be avoided for resuscitation in severe sepsis.
CHF = congestive heart failure; RCT = randomized controlled trial; RR = relative risk.
IMPLICATIONS OF HES IN HEALTHY GRANULOCYTE DONORS For many years macromolecular compounds have been used in collecting granulocyte products to aid in sedimentation of RBCs away from neutrophils to promote purity of the concentrates and enhance the white blood cell (WBC) yield.1,10 Although in the early days of granulocyte collections dextran, gelatin, or HES was used, most centers engaged in collections switched to two main preparations, 6% HES/450/0.7 (hetastarch) or 10% HES/264/0.45 (pentastarch).1,10 The optimal product used was controversial, and early studies claimed comparable efficiencies and granulocyte yields. Kinetic variables and clearance from individuals as well as the toxicity profiles favored pentastarch.44 However, the efficacy of pentastarch for
granulocyte collection was challenged in two clinical studies.11,45 In the former trial in which steroid-treated donors underwent paired granulocyte collections with 6% HES/450/0.7 or 10% HES/264/0.45, in a standardized protocol processing 7 L with 500 mL of the HES preparation, the mean yield was much greater with the former sedimenting agent.11 The enhanced yields by hetastarch were confirmed in several recent studies of granulocyte– colony-stimulating factor (G-CSF)–mobilized donors.46-49 The negative implications for HES use in granulocyte donors are minimal. In general, donors undergo procedures at monthly intervals but this may be more frequent with 1 to 2 weeks as intervals between collections. With each procedure, approximately 7 L of donor blood volume is processed1 with a range in published studies of 7 to 10 L, and 500 mL but usually no more than 750 mL of HES is Volume 55, April 2015 Volume **, ** **TRANSFUSION TRANSFUSION915 5
AMBRUSO
used in the collection. The actual amount the donors receive is approximately 450 to 475 mL as some of the HES remains with the collected granulocytes. This is less than the usual dose given to critically ill patients as a volume expander in the previously reviewed studies. Even if multiple collections were performed with up to two or three donations in a week, the risk for renal problems is very low in the donor population After single apheresis procedures, levels of HES decrease rapidly with a half-life up to 3.8 days and a longer time (6-7 weeks) for complete elimination.11,12 With multiple procedures, the pattern is similar, but persistence of trace amounts of HES may occur up to 72 weeks. Although long-term effects are not precisely known, it seems less likely they will occur on the basis of dose administered and excretion the majority of the drug. In studies of granulocyte donors, the side effects reported with administration of HES, separate from any adverse events related to mobilization with steroids and/or G-CSF, are minimal.11,44-52 Most series have documented pruritus as a complication in a low number of donors (up to 6% in one series, not described in others) and collections (0.1%-0.7%). Consequences related to HES effects on volume or renal dysfunction have not been documented, and no incidents of mortality have been described. A postmarketing review of COBE Spectra system complaints over 10 years from 2001 to 2010 covering approximately 40,000 granulocyte collections did not reveal any serious adverse donor events or deaths (TerumoBCT Spectra Optia apheresis system, personal communications from a clinical evaluation report, 2012). This included serious events related to HES used in the collections. The search revealed only minor device malfunctions. Other adverse reactions (e.g., anaphylaxis, bleeding complications, or others) are either very rare (
