13) Kozek-Langenecker_2008

7/28/2019 13) Kozek-Langenecker_2008

1/9

The Effects of Hydroxyethyl Starch 130/0.4 (6%) onBlood Loss and Use of Blood Products in Major Surgery:

A Pooled Analysis of Randomized Clinical Trials

Sibylle A. Kozek-Langenecker*

Cornelius Jungheinrich**

Wilhelm Sauermann

Philippe Van der Linden

BACKGROUND: The effects of different types of hydroxylethyl starch (HES) on bloodcoagulation closely depend on their physicochemical properties. HES with lowermolar substitution and a lower in vivo molecular weight interferes relatively littlewith hemostasis and therefore results in lower perioperative blood losses and redblood cell (RBC) transfusion. To test this hypothesis, we analyzed pooled data fromall available studies in major surgery comparing 6% HES 130/0.4 and 6% HES200/0.5 from waxy maize starch.METHODS: Estimated blood loss, drainage loss, calculated blood loss, transfusedblood product volumes, and coagulation variables were examined for 24 h after thestart of surgery. Groups were compared using analysis of variance, evaluatingseveral covariates.RESULTS: Four-hundred-forty-nine patients from seven clinical trials were analyzed,

228 received HES 130/0.4, and 221 received HES 200/0.5. For HES 130/0.4 patients,when compared to HES 200/0.5 patients, the estimated blood loss was reduced by404 mL [P 0.006], drainage loss was 272 mL less [P 0.009], and calculated RBCloss was 149 mL less [P 0.003]. RBC transfusion volumes were also lower for HES130/0.4 by 137 mL [P 0.004]. In the early postoperative phase, HES 130/0.4 wasfound to exert significantly less effect on measures of coagulation, especiallyactivated partial thromboplastin time and von Willebrand factor (antigen andristocetin cofactor), than HES 200/0.5.CONCLUSIONS: Blood loss and transfusion requirements can be significantly reducedin major surgery when using third generation HES 130/0.4 (Voluven) comparedto second generation waxy maize starch HES 200/0.5. Since HES 130/0.4 and HES200/0.5 were found similar regarding volume efficacy in other studies, HES130/0.4 is recommended in this clinical setting.(Anesth Analg 2008;107:38290)

Maintaining adequate circulating blood volumeand tissue perfusion is important during major sur-gery. Both crystalloids and colloids can be used butmay influence coagulation beyond hemodilution.1,2

Crystalloids have been reported to have pro-coagulantproperties at low degrees of hemodilution.2 Amongcolloids, hydroxyethyl starch (HES) products havepotential effects on clot formation,35 humoral coagu-lation factors,6 platelet function,711 and clot polymer-ization.12,13 These effects have been reviewed14 and

depend on the pharmacokinetic properties of thespecific HES type used, which in turn determine theHES plasma concentrations over time, present orabsent plasma accumulation, in vivo molecular weight(Mw), and maximum doses.1517

Since the first HES generation (hetastarch, labelede.g. as HES 450/0.7, 550/0.7, 670/0.75), new formula-tions have been developed to improve the safetyprofile of HES, especially regarding coagulation. Untilrecently, pentastarch (HES 200/0.5), a second genera-tion HES, was used as a standard in intravascularvolume therapy with artificial colloids in many Euro-pean countries for about 20 yr. HES 130/0.4 is thelatest generation of the commercially available HESsolutions. It is characterized by a mean in vitro Mw of130,000 20,000 Da, a molar substitution (MS) of 0.4,and a C2/C6 ratio of about 9:1. The low MS is theprimary determinant of increased metabolic degrada-tion, the in vitro Mw playing a less important role. Theincreased C2/C6 ratio partially offsets the effect of thereduced Mw, since a high C2/C6 ratio decreases the rate

of hydrolysis by

-amylase. Such a HES solution istherefore expected to affect coagulation less and reduce

From the *Department of Anesthesiology, General IntensiveCare and Pain Control, Vienna Medical University, Vienna, Austria;**Medical Affairs, Fresenius Kabi, Bad Homburg, Germany;DATAMAP GmbH (Biostatistics Institute), Freiburg, Germany;and Department of Anesthesiology, CHU Brugmann-HUDERF,Brussels, Belgium.

Accepted for publication April 15, 2008.

Supported by Fresenius Kabi.

Address of correspondence and reprint requests to Sibylle Kozek-Langenecker, Department of Special Anesthesiology and Pain Man-agement, Vienna Medical University, Wahringer Gurtel 18-20, 1090Vienna, Austria. Address e-mail to [email protected].

Copyright 2008 International Anesthesia Research Society

DOI: 10.1213/ane.0b013e31817e6eac

Vol. 107, No. 2, August 2008382


2/9

blood loss and blood transfusion when compared toolder HES solutions, such as pentastarch (HES 200/0.5),characterized by a higher MS and Mw.

The purpose of this pooled analysis was to compareHES 130/0.4 (Voluven) and HES 200/0.5 (HAES-steril) from waxy maize starch in terms of periopera-tive blood loss, allogeneic blood product requirementsand coagulation variables in patients undergoing majorsurgical procedures.

METHODS

Studies and Patients

All prospective, randomized, clinical studies compar-ing 6% HES 130/0.4 (Voluven) and the control solution,6% HES 200/0.5 (HAES-steril both Fresenius Kabi, BadHomburg, Germany) have been evaluated. Only trialswith surgical indications were considered, since no

blood loss in either group was to be expected insituations of hypervolemic treatment, e.g., for suddenhearing loss or stroke. Published and unpublishedstudies were included only if individual patient datacould be retrieved.

We used the following search strategy (Fig. 1): 293publications were identified after consolidation of the

EMBASE and MEDLINE search results using thesearch terms hydroxyethyl starch, hydroxyethylstarch,

HES, an(a)emia, orthop(a)edic, surgery, volume re-placement, blood loss, cardiopulmonary bypass, sub-stitution, and coronary artery bypass. The time periodwas limited from 1988 to 2006 (wk 50). No language orstudy design constrains were adopted in conductingthe searches. Relevant studies were also included bymanual searching for unpublished studies, e.g., allstudies used for international regulatory purposes,meeting abstracts.

Two-hundred-three articles for more detailed evalu-ation were available after exclusion of: 1) reviews, casereports, letters, and conference papers as defined by thedatabases, 2) non-human studies, in vitro, and ex vivostudies still contained in the results despite limitingthe search to human studies only, 3) case reports andreviews contained in the results which were notclearly defined as such in the publication type of thesearched databases, 4) pharmacokinetic, pharmacody-namic studies, studies with volunteers, 5) retrospec-tive studies, 6) meta-analyses, and 7) studies usingstarches other than HES.

The remaining 129 articles were further checked ifthey provided comparative data for HES 130/0.4 andHES 200/0.5. Other starch sources (one study usingpotato starch) were excluded because of the pharma-cological differences and the products not being bio-equivalent to starch products. Nine articles remainedfor further consideration. Of these, two studies had to

be excluded because they were not at least partiallyblinded, and no data on blood loss of individualpatients were available.

The manual search for unpublished data complyingwith the search profile yielded one study performed

for regulatory purposes (HS-1324-DE) that was in-cluded in the pooled analysis. Consequently, datafrom seven studies were finally included in the anal-ysis.1722 All of these studies were at least partially

blinded or even fully double-blind (all five studieswith preplanned equal colloid volumes in bothgroups).

Statistical AnalysisThe pooled analysis was based on the original

single patient data of the seven studies. In contrast, ameta-analysis is only based on mean values and

standard deviations for the single studies, mostlytaken from publications. A pooled analysis is superiorto a meta-analysis, since more information is availableand since it allows analyzing the effect of explanatoryvariables on the response variables. Further, it allowsone to define response variables and explanatoryvariables consistently for all studies. Differences be-tween studies can be accounted for and analyzed byincluding a study in the statistical model. Responsevariables were blood loss, drainage loss, calculatedred blood cell (RBC) loss, transfused RBC volume,transfused platelets, transfused fresh frozen plasma,

and coagulation variables during the first 24 h afterthe start of surgery. Transfused RBC volume was

Figure 1. Search strategy. *Reviews, Case Reports, Letters,Conference Papers as defined by the data bases; non-human,in vitro, ex vivo, case reports, reviews (not pre-defined in thedata bases); pharmacokinetics, pharmacodynamics, volun-teers, retrospective studies, meta-analyses. **Studies not usinghydroxyethyl starch (HES) made of waxy maize starch, controldrug other than HES, comparison other than HES 130/0.4versus HES 200/0.5, uncontrolled studies. ***No perioperativeblood loss or no blood loss data; more than one publication ofthe same study population.

Vol. 107, No. 2, August 2008 2008 International Anesthesia Research Society 383


3/9

defined as the sum of all transfused RBC volumesmultiplied by their respective hematocrit (HCT),which was either contained in the database or wasassumed to be 0.7 for packed RBC and 0.6 for salvaged

blood. Calculated RBC loss, using a modification ofthe method of Mercuriali and Inghilleri,23 was definedas estimated blood volume (HCT0 hHCT24 h) transfused RBC volume, whereby blood volume was

estimated by the formula of Nadler et al.24 Estimatedblood loss was defined as the sum of objective losses,e.g., via drainage and swabs plus clinically estimatedadditional losses. Several possible explanatory vari-ables were analyzed to assess their influence on theresponse variables.

Results were reported as mean sd (SD), median,range and number of observations (N) (quantitativevariables) or as absolute and relative frequencies(qualitative variables).

The statistical analysis was performed by an ana-lysis of variance with treatment group, study, and

explanatory variables as effects. During the analysisplanning phase, 22 possible explanatory variableswere defined, each of which divided the total patientsample into subgroups. Many of these explanatoryvariables either led to small subgroups, were con-founded with one of the other explanatory variables,or were not considered true explanatory variables inthe sense that their outcome was measured after startof treatment (i.e., was a response rather than anexplanatory variable). These variables were excludedfrom the analysis. The model which contained allremaining explanatory variables was termed full

model and included treatment group, study and theexplanatory variables age, sex, duration of surgery,aspirin administration, and aprotinin at day of sur-gery. Interactions were investigated for significantvariables to assess the uniformity of the treatmenteffect over subgroups defined by the explanatoryvariables. Single models including only one ex-planatory variable and additional treatment groupand study were also investigated in order to assess thegoodness of fit of the full model and possible con-founding of explanatory variables. Statistical resultspresented here are based on the full model.

Data were analyzed using SAS

Version 8.2, SASInstitute Inc., Cary, NC.

RESULTS

Seven clinical trials fulfilled the pre-defined selec-tion criteria and are listed in Table 1. Four-hundred-forty-nine patients were therefore analyzed in apooled fashion, 228 in the HES 130/0.4 and 221 in theHES 200/0.5 group. All seven studies were prospec-tively randomized and conducted in settings wherehigh surgical blood losses were anticipated, specifi-cally in cardiac surgery (four studies), major orthope-dic surgery (two studies), and urologic surgery (onestudy). All but two studies were double-blind and hadsimilar maximum doses of study colloids. Studies

HS-1324-DE and Kasper et al.

22

were blinded until adose of 33mL/kg body weight, and unblinded only

Table 1. Studies Included in the Pooled Analysis

Study number SettingNumber of patients

HES 130/0.4Number of patients

HES 200/0.5

HS-1313-DE (Jungheinrich17) Orthopedic surgery 26 26HS-1314-NL (Gallandat18) Cardiac surgery 30 29HS-1318-BE (lckx19) Urologic surgery 20 20HS-1319-FR (Langeron20) Orthopedic surgery 52 48HS-1321-DE (Boldt21) Cardiac surgery 10 10HS-1324-DE Cardiac surgery 31 30

Kasper et al. 200322

Cardiac surgery 59 58Total 228 221

HES hydroxyethyl starch.

Table 2. Demographic Data

VariableHES 130/0.4

(6%)HES 200/0.5

(6%) P

n 228 221Age (yr) 60.3 11.4 62.4 9.9 0.04*Sex (F/M) 74/154 75/146 0.76Weight (kg) 76.7 11.1 75.6 11.0 0.23Height (cm) 169.7 8.8 169.4 9.4 0.64

BMI 26.6 3.4 26.4 3.3 0.34Data are presented as mean SD unless otherwise noted.

BMI Body Mass Index; n number of patients in the pooled analysis; HES hydroxyethylstarch.

P-values are two-sided from Fishers exact test (sex) and an ANOVA with study and treatment

as effects (all other parameters).

* P 0.05.

Table 3. Volumes of HES Solutions Infused

Variable HES 130/0.4 HES 200/0.5 P

n 228 221HES volume (mL) 2,546 1,075 2,104 622 0.01*

2.5 L infused (n

) 129 57% 183 83%2.5 L infused (n) 99 43% 38 17%HES volume (mL/

kg BW)33.2 13.2 28.0 7.9 0.01*

35 mL/kg BW (n) 133 58% 194 88%35 mL/kg BW (n) 95 42% 27 12%

Data are presented as mean SD unless otherwise noted.

Higher volumes per kg body weight of HES 130/0.4 compared to HES 200/0.5 were allowed

by several study protocols, which expectedly led to a higher exposure to HES 130/0.4.

HES hydroxyethyl starch; BW body weight; n number of patients in the pooledanalysis.

P-values are two-sided from an ANOVA with study and treatment as effects.

* P 0.05.

384 HES 130/0.4 Reduces Bleeding ANESTHESIA & ANALGESIA


4/9

after the colloid dose of 33 mL/kg was reached. In

both studies, the HES 130/0.4 group could then re-ceive more study colloid until 50 mL/kg, whereas theHES 200/0.5 control groups had to switch to gelatin.

Demographic variables are summarized in Table 2.There was no significant difference in baseline char-acteristics between the two groups of patients except aslightly higher age in the control group. Table 3displays the HES volumes given in both groups. Dueto the two protocols which mandated unequal HESvolumes, the number of patients receiving largeramounts of HES (2.5 L or 35 mL/kg) was aboutthreefold in the HES 130/0.4 group (99 vs 38, and 95 vs

27). Therefore, mean exposure to HES 130/0.4 studysolution was higher than to the HES 200/0.5 controlsolution.

Table 4 shows raw means sd for blood loss andtransfusion variables, stratified by type of surgery.Results of the pooled statistical analysis are displayedin Figures 25, always referring to the HES 130/0.4-HES 200/0.5 difference.

Perioperative blood loss was significantly less in theHES 130/0.4 group than in the HES 200/0.5 group. Forthe clinically evaluated blood loss, the mean estimateddifference [95% confidence interval] derived from ana-

lysis of variance was

404 [

689;

119] mL (P

0.006,Fig. 2). For drainage loss, the mean estimated difference

was 272 [474; 70] mL (P 0.009; Fig. 3). For the

calculated RBC loss, the mean estimated difference was149 [247; 50] mL (P 0.003; Fig. 4).

Figure 2. Estimated blood loss in pooled analysis and insingle studies, differences hydroxyethyl starch (HES)130/0.4HES 200/0.5. Negative values mean less blood lossin the HES 130/0.4 group. Total: Least squares estimates and95% confidence interval based on Analysis of Variance fullmodel; single study strata: raw means and 95% confidenceintervals. For information on single studies refer to Table 1.

Table 4. Blood Loss and Transfusions by Type of Surgery

Variable, type of surgery n HES 130/0.4 n HES 200/0.5 P

Estimated blood loss (mL)Cardiac 128 1601 1714 127 1603 1486 0.77Orthopedic 78 2415 1430 74 3112 2295 0.01*Urologic 20 1971 748 20 3340 3861 0.25Total 226 1915 1597 221 2266 2213 0.01*

Drainage loss (mL)Cardiac 118 930 539 117 1050 850 0.05*Orthopedic 78 779 639 74 980 825 0.07Urologic 20 1971 748 20 3340 3861 0.25Total 216 972 679 211 1242 1565 0.01*

Calculated RBC loss (mL)Cardiac 128 860 409 122 930 492 0.09Orthopedic 75 798 514 73 989 842 0.08Urologic 20 761 241 20 1109 840 0.20Total 223 830 436 215 967 664 0.01*

Transfused RBC volumes(mL)

Cardiac 128 383 329 127 464 491 0.06Orthopedic 78 616 505 74 786 813 0.13Urologic 20 12 55 20 280 775 0.27Total 226 430 421 221 555 663 0.01*

Transfused platelets (mL)Cardiac 128 8.6 46.6 127 12.6 72.4 0.39Orthopedic 78 0.0 0.0 74 15.9 98.0 0.39Urologic 20 0.0 0.0 20 10.0 44.7 0.53Total 226 4.9 35.3 221 13.5 79.8 0.21

Fresh frozen plasma (mL)Cardiac 128 116 315 127 192 594 0.13Orthopedic 78 226 438 74 156 377 0.35Urologic 20 40 179 20 80 358 0.82Total 226 147 358 221 170 511 0.47

Data are presented as mean SD unless otherwise noted.

HES hydroxyethyl starch; RBC red blood cell; n number of patients for the respective parameter.

P-values are two-sided, based on an ANOVA (full model).

* P 0.05.



5/9

In line with the results regarding RBC loss, trans-fused RBC volume was also significantly lower in theHES 130/0.4 group, with a mean estimated differenceof137 [231; 43] mL (P 0.004; Fig. 5). Differ-ences in calculated RBC loss and transfused RBCvolume between the two HES groups appeared toincrease with the duration of the surgical procedure.Differences in the transfused volume of platelets (7.4[19.0; 4.1] mL, P 0.21) and FFP (30 [113; 53]mL, P 0.47) were not significant.

In the early postoperative period, the coagulationprofile also appeared different between the two

groups (Table 5). Significant differences were foundfor the early postoperative phase at 48 h after end ofsurgery. Activated partial thromboplastin time (P 0.02) was shorter in the HES 130/0.4 group, vonWillebrand factor-ristocetin cofactor (P 0.04) andvon Willebrand factor-antigen (P 0.04) were signifi-cantly higher for HES 130/0.4. Mean Factor VIII wasalso higher for the HES 130/0.4 group but did not

reach significance in this pooled analysis. Plateletcounts were slightly but significantly higher for HES130/0.4 patients at this time point and at the end ofsurgery (P 0.03). Quick values were also slightlyhigher at 24 h in the HES 130/0.4 group.

DISCUSSION

Our pooled analysis of the different studies havingcompared HES 130/0.4 and HES 200/0.5 in majorsurgical procedures demonstrated that the use of HES130/0.4 is associated with a significant reduction inperioperative blood loss, resulting in a decrease in

allogeneic blood transfusion equivalent to about oneRBC unit per patient. A pooled analysis of the avail-able single patient data was possible, and not merely ameta-analysis of published data. Results were remark-ably homogeneous between studies and consistent forthe different variables, with a reduced calculated RBCloss of 149 mL with HES 130/0.4, almost equal to adecreased transfused RBC volume of 137 mL in thesame group during the first 24 h after the beginning ofsurgery. The duration of surgery appeared to be animportant factor, since the differences between thetwo groups became larger with increased surgery

time. Univariate analyses, as described in the Methodssection, yielded results consistent with the full model.

Figure 3. Drainage loss in pooled analysis and in singlestudies, differences hydroxyethyl starch (HES) 130/0.4 HES200/0.5. Negative values mean less drainage loss in the HES130/0.4 group. Total: Least squares estimates and 95%confidence interval based on Analysis of Variance full

model; single study strata: raw means and 95% confidenceintervals. For information on single studies refer to Table 1.

Figure 4. Calculated red blood cell loss in pooled analysisand in single studies, differences HES 130/0.4HES 200/0.5.Negative values mean less red blood cell loss in the HES130/0.4 group. Total: Least squares estimates and 95%confidence interval based on Analysis of Variance fullmodel; single study strata: raw means and 95% confidence

intervals. For information on single studies refer to Table 1.

Figure 5. Transfused red blood cell volume in pooled anal-ysis and in single studies, differences hydroxyethyl starch(HES) 130/0.4HES 200/0.5. Negative values mean lowertransfused red blood cell volumes in the HES 130/0.4 group.Total: least squares estimates and 95% confidence interval

based on Analysis of Variance full model; single studystrata: raw means and 95% confidence intervals. For infor-mation on single studies refer to Table 1.



6/9

Therefore, the chosen full model can be judged asrobust for this pooled analysis.

Differences between HES 130/0.4 and 200/0.5 ap-pear larger than those reported in a previous meta-analysis comparing high in vivo Mw starches (HES450/0.7), second generation HES 200/0.5, and albu-min in cardiac surgery.25 Wilkes et al.25 failed to showa significant difference between HES 200/0.5 andalbumin regarding drainage loss (0 included in the

95% CI), but found an overall significant difference of93 mL (789 487 mL vs 693 350 mL) in favor ofalbumin compared to the HES products they exam-ined. Although blood product use could not be quan-tified overall in that meta-analysis, blood producttransfusion was lower in the albumin group in abouthalf of the trials included (8/15).

Our pooled analysis included three studies per-formed in cardiac surgery in which drainage loss wasreported. In the study in which comparable amountsof study colloids were given (HS-1314-DE), drainageloss was 520 mL less in the HES 130/0.4 group

compared to HES 200/0.5 (Fig. 3). In the two othercardiac studies (HS-1324-DE and 22), noninferiority

of blood loss in the high dose group was provenwithin these trials, despite a preplanned 50% higherdose of 50 mL/kg HES 130/0.4 compared to 33 mL/kgHES 200/0.5 plus added gelatin. If these high doseHES trials with unequal doses in both groups wereremoved from the analysis, differences in blood lossand transfusion requirements would have been evenhigher than observed for the full set of studies.

Other studies not covered by our pooled analysis

directly compared HES 130/0.4 with crystalloids. Inmajor abdominal surgery, Lang et al. reported similarblood loss and use of blood products for bothgroups.26 In another study, Boldt et al.1 showed simi-lar blood loss until 24 h for HES 130/0.4 (1180 250mL) and Ringers lactate solution (1210 350 mL), buthigher blood loss for balanced HES 550/0.7(hetastarch, 1680 270 mL). A similar relation wasseen for the use of blood products. Modified throm-

belastography was more impaired after hetastarch.A cardiac surgery study by Van der Linden et al.27

compared HES 130/0.4 (n 64; dose 49 18 mL/kg)

with modified fluid gelatin (n

68; dose 49

15mL/kg). Measured blood loss was similar in both

Table 5. Coagulation Variables

Variable, time HES 130/0.4 HES 200/0.5HES 130/0.4- HES 200/0.5

95% CI P

aPTT (s)End of surgery (EOS) 46.1 2.7 48.5 2.7 2.4 6.1; 1.2 0.1948 h after EOS 34.8 1.2 38.3 1.2 3.5 6.5; 0.6 0.02*24 h after start 38.8 1.7 39.7 1.8 0.9 3.3; 1.5 0.47

Quick (%) (prothrombin time)End of surgery (EOS) 71.9 2.3 72.4 2.3 0.5 3.2; 2.2 0.71

48 h after EOS 77.3 1.6 73.7 1.6 3.6 0.2; 7.5 0.0624 h after start 82.0 1.8 79.3 1.9 2.7 0.3; 5.2 0.03*

Thombin time (s)End of surgery (EOS) 14.8 1.3 16.2 1.3 1.4 4.1; 1.2 0.2948 h after EOS 18.4 2.1 20.2 2.2 1.8 7.0; 3.4 0.4924 h after start 13.6 1.2 13.4 1.3 0.2 2.5; 2.9 0.87

von Willebrand factor (ristocetincofactor activity, %)

End of surgery (EOS) 52.1 28.2 46.5 27.2 5.6 12.3; 23.6 0.5448 h after EOS 124.6 6.6 107.4 6.8 17.2 1.0; 33.4 0.04*24 h after start 197.7 37.7 164.0 35.4 33.7 0.7; 68.1 0.06

von Willebrand factor (antigen, %)End of surgery (EOS) 100.4 4.4 95.9 4.5a 2.1 10.2; 14.5 0.7348 h after EOS 111.8 6.2 96.1 6.4 15.7 0.6; 30.8 0.04*24 h after start 139.2 30.7 127.1 28.1 12.1 17.7; 41.9 0.42

Factor VIII C (chromogenic method, %)End of surgery (EOS) 80.5 11.0 80.5 10.4 0.0 9.0; 9.0 1.0048 h after EOS 104.7 6.2 90.7 6.3 14.0 0.7; 28.7 0.0624 h after start 115.2 17.7 116.8 16.3 1.6 17.8; 14.7 0.85

Fibrinogen (g/L)End of surgery (EOS) 1.78 0.10 1.71 0.10 0.08 0.06; 0.21 0.2748 h after EOS 1.85 0.11 1.80 0.11 0.05 0.16; 0.25 0.6424 h after start 3.17 0.16 3.18 0.17 0.01 0.25; 0.22 0.92

Platelets ( times 109/L)End of surgery (EOS) 158 5 150 5 8 1; 15 0.03*48 h after EOS 168 5 155 5 13 2; 25 0.03*24 h after start 166 5 160 5 6 1; 13 0.10

Data are presented as least squares means SE , unless otherwise noted, generated from the ANOVA full model as explained in the methods section.

CI confidence interval; aPTT activated partial thromboplastin time; HES hydroxyethyl starch.a

Estimates SE for this time point from model including treatment and study (because of too low n for full model estimation).* P 0.05.



7/9

groups (HES, 19.4 12.3 mL/kg, gelatin, 19.2 14.5mL/kg), and calculated net RBC loss was also equiva-lent. Neither the number of patients transfused nor thevolume of blood products given differed. This con-firmed earlier evidence in cardiac surgery28 and majorabdominal surgery.29

Direct comparisons of blood loss after slowly me-tabolizable hetastarch (MS 0.7) and HES 130/0.4were performed by Boldt et al.1 and Gandhi et al.30 In

the latter study, patients undergoing major orthopedicsurgery received either hetastarch in saline (n 51) orHES 130/0.4 (n 49) in a double-blind fashion. RBCtransfusion was significantly higher in the hetastarchgroup (13.8 12.9 mL/kg) compared to HES 130/0.4(8.0 6.4 mL/kg).

The coagulation variables reported in our pooledanalysis provide some mechanistic hints to explain thelower blood loss and transfusion requirements ob-served in patients receiving HES 130/0.4 compared tothose receiving HES 200/0.5. Partial thromboplastintime, von Willebrand factor antigen and ristocetin

cofactor activity were significantly higher in the earlypostoperative phase in HES 130/0.4 patients com-pared to HES 200/0.5 patients. In addition, there wasa strong trend in the same direction for Factor VIII.These differences indicated a lower impact of HES130/0.4 on the plasmatic coagulation system thanHES 200/0.5. However, it must be recognized thatglobal coagulation variables alone, such as partialthromboplastin time, Quick (prothrombin time) andthrombin time have a poor sensitivity to predict

bleeding complications.31

Slight but statistically significant differences in

platelet counts (13/nL at 48 h postoperatively) oc-curred after HES 130/0.4 versus HES 200/0.5 admin-istration (Table 5). Since the static test of platelet countdoes not correlate to platelet function, this findingallows no definite conclusion on a potential platelet-associated mechanism of reduced bleeding after HES130/0.4 infusion. Tests of the platelet response toagonists at the cellular level using flow cytometry, andan overall measure of platelet function using theplatelet function analyzer, permitted the detection andcharacterization of HES-induced antiplatelet effects asa function of degradability.7 Higher platelet numbers,

together with better preserved platelet reactivity, mayaccount for the reduced bleeding after HES 130/0.4infusion. Less impairment of the interaction betweenthrombin, factor XIII, and fibrinogen in the presence ofHES 130/0.412,13 may also have resulted in strongerclot polymerization and reduced bleeding.

After infusion of HES 130/0.4, which is rapidlymetabolized, plasma concentrations and in vivo Mware lower compared to less metabolized HES prod-ucts.17 Total body clearance of HES 130/0.4 is about56 times higher than for HES 200/0.5, and 2331times higher than for hetastarch15,32; renal excretion of

HES 130/0.4 is preserved in mild to severe nonanuricrenal impairment.33 The combination of increased

clearance and optimized in vivo Mw of HES 130/0.4 isthought to be responsible for the reduced effects ofHES 130/0.4 on coagulation.15,17 In contrast, hetastarchaccumulates in plasma.34 High persisting concentra-tions of a colloid should be avoided, since they are notexpected to exert a relevant therapeutic effect but mayhave adverse effects. The increased therapeutic safetyindex of HES 130/0.4 compared to less metabolizablestarches has been acknowledged by European Union

countries and by an increasing number of overseasregulatory authorities by changing the former maxi-mum daily dose from 33 to 50 mL/kg body weight perday.35 Waxy maize starch HES 130/0.4 (Voluven) hasrecently been approved by the Food and Drug Admin-istration36 at a maximal daily dose of 50 mL/kg.

The efficacy of plasma volume substitution by HES130/0.4 and HES 200/0.5 has been demonstrated to becomparable in several pivotal trials.17,18,20 The extentand duration of the plasma volume effect of HES130/0.4 were at least as sustained as after hetastarchadministration in volunteers when using the radiola-

beling technique.37 Efficacy in volume substitutiontherapy was also found to be similar betweenhetastarch and HES 130/0.4 during major orthopedicsurgery.30

A different pentastarch product, 10% HES 200/0.5(Hemohes), was associated with higher rates of acuterenal failure and renal replacement therapy than wasa modified Ringers solution containing 45 mmol/L oflactate in critically ill patients in the recent multicenterVISEP trial.38 In that trial, high cumulative doses weregiven up to 21 days (median, 70 mL/kg, range notreported), and the dose limitation for that specific

product of 20 mL

kg1

d1

was significantly ex-ceeded in 100/262 subjects. Ten percent HES 200/0.5administered in patients with contraindicationsagainst the use of HES (e.g., anuric, renal replacementtherapy) may explain this observation. Furthermore,hyperoncotic solutions like HES 200/0.5 (10%) mayproduce renal dysfunction.39 Studies with tetrastarch130/0.4 (6%) which, in contrast to HES 200/0.5 (10%)is isooncotic and which does not accumulate inplasma, did not show a deterioration of renal functioncompared to controls when used in volunteers withmoderate to severe nonanuric renal failure,33 in el-

derly patients undergoing major surgery,40

in cardiacand aortic surgery patients with renal dysfunc-tion4143, and in intensive care patients.44

Limitations of our pooled analysis have to beconsidered when extrapolating the results to clinicalpractice. Heterogeneity in study designs of individualtrials is a common problem of both pooled analysesand meta-analyses. Also, our total number of patientsin the studies is relatively small. Further, crystalloids,additional colloids (such as albumin and gelatin),carrier solutions, as well as transfusion triggers and

blood replacement management were not standard-

ized across individual trials. However, since statisti-cally significant and consistent results were obtained



8/9

despite heterogeneous study designs, the clinical ef-fect size of the investigated variable must be high.From the statistical view, conducting a meta-analysisor pooled analysis is justified in such situations espe-cially if (also unknown) heterogeneity is included inthe mathematical model as a cofactor. From the clini-cal view, heterogeneity in the analyzed trials may

better represent varying clinical practice compared toone study design with one homogenous fluid and

transfusion management.Another limitation of our pooled analysis is that

coagulation tests were done in some studies but not inothers, time points of blood withdrawal were notidentical, and the performed conventional laboratorytests are not sensitive to the pathomechanism ofdilutional coagulopathy. However, the present resultsobtained at similar time slots confirm previous stud-ies313 showing a clear difference between pentastarchand tetrastarch. Considering the observed differencesin conventional laboratory tests as indicators, we

believe that the conclusion is justified to identify the

lower antihemostatic side effects of tetrastarch as thereason for reduced perioperative blood loss. Never-theless, further studies directly comparing HES130/0.4 and HES 200/0.5 have to verify this hypoth-esis by using point-of-care tests sensitive to plateletfunction, clot formation, and clot polymerization, aswell as to the multifactorial nature of bleeding associ-ated with trauma and major cardiac, orthopedic andurologic surgery.

In conclusion our data show that blood loss andtransfusion requirements can be significantly reducedin patients undergoing major surgery when using

third generation HES 130/0.4 (Voluven

) compared tosecond generation starch HES 200/0.5. Since HES130/0.4 and HES 200/0.5 were found similar regard-ing volume efficacy in other studies, HES 130/0.4should be preferred to less rapidly metabolizable HESsolutions in prevention and treatment of perioperativehypovolemia, especially if large volumes are required.

ACKNOWLEDGMENTS

Dr. Sauermann is head of DATAMAP, a Contract Re-search Organization providing biometrical services. His in-stitutions received funding for biometrical services related toseveral clinical studies of Fresenius Kabi, all performedaccording to GCP.

Fresenius Kabi, Bad Homburg, Germany, manufacturedboth materials (6% HES 130/0.4 and 6% HES 200/0.5)discussed in the manuscript.

REFERENCES

1. Boldt J, Haisch G, Suttner S, Kumle B, Schellhaass A. Effects of anew modified, balanced hydroxyethl starch preparation (Hex-tend) on measures of coagulation. Br J Anaesth 2002;89:7228

2. Ruttmann TG, James MF, Aronson I. In vivo investigation intothe effects of haemodilution with hydroxyethyl starch (200/0.5)and normal saline on coagulation. Br J Anaesth 1998;80:6126

3. Jamnicki M, Bombeli T, Seifert B, Zollinger A, Camenzind V,Pasch T, Spahn DR. Low- and medium-molecular-weight hy-

droxyethyl starches: comparison of their effects on blood coagu-lation. Anesthesiology 2000;93:12317

4. Innerhofer P, Fries D, Markreiter J, Klingler A, Kuhlbacher G,Wachter B, Oswald E, Salner E, Frischhut B, Schobersberger W.The effects of perioperatively administered colloids and crystal-loids on primary platelet-mediated hemostasis and clot forma-tion. Anaesth Analg 2002;95:858 65

5. Omar MN, Shouk TA, Khaleq MA. Activity of blood coagula-tion and fibrinolysis during and after hydroxyethyl starch (HES)colloidal volume replacement. Clin Biochem 1999;32:26974

6. Stump DC, Strauss RG, Henriksen RA, Petersen RE, Saunders R.Effects of hydroxyethyl starch on blood coagulation, particu-larly factor VIII. Transfusion 1985;25:34954

7. Stogermuller B, Stark J, Willschke H, Felfernig M, Horauf K,Kozek-Langenecker S. The effects of hydroxyethyl starch 200 kDon platelet function. Anesth Analg 2000;91:8237

8. Gamsjager T, Gustorff B, Kozek-Langenecker S. Effects of hy-droxyethyl starches on intracellular calcium in platelets. AnesthAnalg 2002;95:8669

9. Deusch E, Gamsjager T, Kress H, Kozek-Langenecker S. Bindingof hydroxyethyl starch molecules to the platelet surface. AnesthAnalg 2003;97:6803

10. Franz A, Braunlich P, Gamsjager T, Felfernig M, Gustorff B,Kozek-Langenecker S. The effects of hydroxyethyl starches ofvarying molecular weights on platelet function. Anesth Analg2001;92:14027

11. Huttner I, Boldt J, Haisch G, Suttner S, Kumle B, Schulz H.Influence of different colloids on molecular markers of haemo-stasis and platelet function in patients undergoing major ab-

dominal surgery. Br J Anaesth 2000;85:4172312. Fries D, Innerhofer P, Reif C, Streif W, Klingler A, Schobersberger W,Velik-Salchner C, Friesenecker B. The effect of fibrinogen sub-stitution on reversal of dilutional coagulopathy: an in vitromodel. Anesth Analg 2006;102:34751

13. Nielsen VG. Effects of PentaLyte and Voluven hemodilution onplasma coagulation kinetics in the rabbit: role of thrombin-fibrinogen and factor XIII-fibrin polymer interactions. ActaAnaesthesiol Scand 2005;49:126371

14. Kozek-Langenecker SA. Effects of hydroxyethyl starch solutionson hemostasis. Anesthesiology 2005;103:65460

15. Jungheinrich C, Neff TA. Pharmacokinetics of hydroxyethylstarch. Clin Pharmacokinet 2005;44:68199

16. Treib J, Haass A, Pindur G, Seyfert UT, Treib W, Grauer MT,Jung F, Wenzel E, Schimrigk K. HES 200/0.5 is not HES 200/0.5 Influence of the C2/C6 hydroxyethylation ratio of hydroxyethyl

starch (HES) on hemorheology, coagulation and eliminationkinetics. Thromb Haemost 1995;74:1452617. Jungheinrich C, Sauermann W, Bepperling F, Vogt NH. Volume

efficacy and reduced influence on measures of coagulationusing hydroxyethyl starch 130/0.4 (6%) with an optimised invivo molecular weight in orthopaedic surgery a randomised,double-blind study. Drugs R D 2004;5:19

18. Gallandat Huet RC, Siemons AW, Baus D, van Rooyen-ButijnWT, Haagenaars JA, van Oeveren W, Bepperling F. A novelhydroxyethyl starch (Voluven) for effective perioperativeplasma volume substitution in cardiac surgery. Can J Anaesth2000;47:120715

19. Ickx BE, Rigoleit M, Van der Linden PJ. Cardiovascular andmetabolic response to acute normovolemic anemia. Anesthesi-ology 2000;93:10116

20. Langeron O, Doelberg M, Ang ET, Bonnet F, Capdevila X,

Coriat R. Voluven, a lower substituted novel hydroxyethylstarch (HES 130/0.4), causes fewer effects on coagulation inmajor orthopedic surgery than HES 200/0.5. Anesth Analg2001;92:85562

21. Boldt J, Lehmann A, Rompert R, Haisch G, Isgro F. Volumetherapy with a new hydroxyethyl starch in cardiac surgicalpatients before cardiopulmonary bypass. J Cardiothorac VascAnesth 2000;14:2648

22. Kasper SM, Meinert P, Kampe S, Gorg C, Geisen C, Mehlhorn U,Diefenbach C. Large-dose hydroxyethyl starch 130/0.4 does notincrease blood loss and transfusion requirements in coronaryartery bypass surgery compared with hydroxyethyl starch200/0.5 at recommended doses. Anesthesiology 2003;99:427

23. Mercuriali F, Inghilleri G. Proposal of an algorithm to help thechoice of the best transfusion strategy. Curr Med Res Opin1996;13:46578

24. Nadler SB, Hidalgo JU, Bloch T. Prediction of blood volume innormal human adults. Surgery 1962;51:22432



9/9

25. Wilkes MM, Navickis RJ, Sibbald WJ. Albumin versus hydroxy-ethyl starch in cardiopulmonary bypass surgery: a meta-analysis of postoperative bleeding. Ann Thorac Surg 2001;72:52734

26. Lang K, Boldt J, Suttner S, Haisch G. Colloids versus crystalloidsand tissue oxygen tension in patients undergoing major ab-dominal surgery. Anesth Analg 2001;93:4059

27. Van der Linden PJ, De Hert SG, Deraedt D, Cromheecke S, DeDecker K, De Paep R, Rodrigus I, Daper A, Trenchant A.Hydroxyethyl starch 130/0.4 versus modified fluid gelatin forvolume expansion in cardiac surgery patients: the effects onperioperative bleeding and transfusion needs. Anesth Analg

2005;101:6293428. Haisch G, Boldt J, Krebs C, Suttner S, Lehmann A, Isgro F.

Influence of a new hydroxyethylstarch preparation (HES 130/0.4) oncoagulation in cardiac surgical patients. J Cardiothorac VascAnesth 2001;15:31621

29. Haisch G, Boldt J, Krebs C, Kumle B, Suttner S, Schulz A. Theinfluence of intravascular volume therapy with a new hydroxy-ethyl starch preparation (6% HES 130/0.4) on coagulation inpatients undergoing major abdominal surgery. Anesth Analg2001;92:56571

30. Gandhi SD, Weiskopf RB, Jungheinrich C, Koorn R, Miller D,Shangraw RE, Prough DS, Baus D, Bepperling F, Warltier DC.Volume replacement therapy during major orthopedic surgeryusing Voluven (hydroxyethyl starch 130/0.4) or hetastarch.Anesthesiology 2007;106:11207

31. Ciavarella D, Reed RL, Counts RB, Baron L, Pavlin E, Heimbach

DM, Carrico CJ. Clotting factor levels and the risk of diffusemicrovascular bleeding in the massively transfused patient. Br JHaematol 1987;67:3658

32. Waitzinger J, Bepperling F, Pabst G, Opitz J, Muller M, Baron JB.Pharmacokinetics and tolerability of a new hydroxyethyl starch(HES) specification [HES(130/0.4)] after single-dose infusion of6% or 10% solutions in healthy volunteers. Clin Drug Investig1998;16:15160

33. Jungheinrich C, Scharpf R, Wargenau M, Bepperling F, Baron JF.The pharmacokinetics and tolerability of an intravenous infu-sion of the new hydroxyethyl starch 130/0.4 (6%, 500 mL) inmild-to-severe renal impairment. Anesth Analg 2002;95:54451

34. Wilkes NJ, Woolf RL, Powanda MC, Gan TJ, Machin SJ, Webb A,Mutch M, Bennett-Guerrero E, Mythen M. Hydroxyethyl starch in

balanced electrolyte solution (Hextend)-pharmacokinetic andpharmacodynamic profiles in healthy volunteers. Anesth Analg

2002;94:53844

35. Swedish Product Information for Voluven. Available athttp://www.lakemedelsverket.se/upload/SPC_PIL(Pdf/humspc/Voluven%20for%20infusion.pdf. Assessed on June 8, 2007

36. http://www.fda.gov/bbs/topics/NEWS/2007/NEW01765.html.Accessed on May 26, 2008.

37. James MF, Latoo MY, Mythen MG, Mutch M, Michaelis C,Roche AM, Burdett E. Plasma volume changes associated withtwo hydroxyethyl starch colloids following acute hypovolaemiain volunteers. Anaesthesia 2004;59:738 42

38. Brunkhorst FM, Englel C, Bloos F, Meier-Hellmann A, RagallerM, Weiler N, Moerer O, Gruendling M, Oppert M, Grond S,

Olthoff D, Jaschinski U, John S, Rossaint R, Welte T, Schaefer M,Kern P, Kuhnt E, Kiehntopf M, Hartog C, Natanson C, LoefflerM, Reinhart K. Intensive insulin therapy and pentastarch resus-citation in severe sepsis. N Engl J Med 2008;358:12539

39. Bartels C, Hadzik B, Abel M, Roth B. Renal failure associatedwith unrecognized hyperoncotic states after pediatric heartsurgery. Intensive Care Med 1996;22:4924

40. Boldt J, Brenner T, Lehmann A, Lang J, Kumle B, Werling C.Influence of two different volume replacement regimens onrenal function in elderly patients undergoing cardiac surgery:comparison of a new starch preparation with gelatin. IntensiveCare Med 2003;29:7639

41. Boldt J, Brosch C, Ducke M, Papsdorf M, Lehmann A. Influenceof volume therapy with a modern hydroxyethylstarch prepara-tion on kidney function in cardiac surgery patients with com-promised renal function: a comparison with human albumin.Crit Care Med 2007;35:27406

42. Mahmood A, Gosling P, Vohra RK. Randomized clinical trialcomparing the effects on renal function of hydroxyethyl starchor gelatine during aortic aneurysm surgery. Br J Surg 2007;94:42733

43. Godet G, Lehot J, Janvier G, Steib A, de Castro V, Coriat P.Safety of HES 130/0.4 (Voluven) in patients with prcoperativerenal dysfunction undergoing abdominal aortic surgery: a pro-spective, randomized, controlled, parallel-group multicentretrial. Eur J Anaesthesiol 2008;19

44. Neff TA, Doelberg M, Jungheinrich C, Sauerland A, Spahn DR,Stocker R. Repetitive large-dose infusion of the novel hydroxy-ethyl starch 130/0.4 in patients with severe head injury. AnesthAnalg 2003;96:14539


13) Kozek-Langenecker_2008

Documents

Transcript of 13) Kozek-Langenecker_2008