The risks associated with transfusion of homologous blood products are well known, and in many hospitals a blood-sparing strategy has been developed . The consumption of albumin solutions is also being reduced because of high cost and limited availability . Artificial colloids represent an alternative for replacing intraoperative blood loss (certainly if >or=to 2000 mL).
Six percent hydroxyethyl starch (HES; molecular weight [MW] 200,000; substitution ratio 0.4-0.55) (Haes-Steril Registered Trademark, Fresenius AG, Bad Homburg, Germany) has a medium to short intravascular volume effect [plateau 3-4 h ] but slightly decreases von Willebrand factor function and alters fibrin formation in usual clinical doses . Three per cent modified fluid gelatin (GEL; MW 35,000); (Geloplasma Registered Trademark, Institut Merieux Benelux, Brussels, Belgium) is inexpensive, has a short volume effect [plateau 1-2 h ], does not influence coagulation other than by dilution , but causes several more serious allergic reactions than HES . Originally developed in the United States , but rarely used there currently, GEL is a degradation product of collagen, modified by the addition of succinic acid anhydride to overcome problems of gelation, high viscosity, and allergic side effects .
Most studies examine only the effects of 500-1000 mL of artificial colloid [10-12]. To further reduce the infusion of albumin, we measured the effects on intravascular volume and coagulation of approximate 2000 mL HES or GEL before and during total hip replacement. To allow quantification of the weaker intravascular volume effect of GEL, and a possible more pronounced hypocoagulation with HES, measurements of plasma volume and artificial colloid concentration in serum and urine were added to the more standard assessments such as hemodynamic variables, fluid balance, and coagulation tests.
The study protocol was approved by the Ethical Committee for Research in patients of the Katholieke Universiteit Leuven and written consent was obtained from all patients. Forty-two patients, scheduled for primary total hip replacement, were included in the study. Inclusion criteria were: ASA physical status I or II, age between 40 and 80 yr, and body weight between 50 and 90 kg. Exclusion criteria were hematocrit (Hct) <35%, preexisting coagulopathy, abnormal preoperative screening tests (prothrombin time <70%, activated partial thromboplastin time >40 s, platelet count <100,000/micro Liter), salicylic acid or analogs taken during the last 8 days before surgery, serum creatinine >1.5 mg/dL , and known allergy to artificial colloids. The patients were allocated randomly to two groups. One group received 6% low molecular weight HES. The other group received 3% GEL. The reported colloid osmotic pressures after stabilization in vitro are 34 mm Hg for 6% HES and approximate 13 mm Hg for 3% GEL (membrane cutoff 10,000 daltons) .
Anesthesia was induced with droperidol, fentanyl, thiopental, and vecuronium intravenously and maintained by mechanical lung ventilation (PCO2 35-40 mm Hg), inhalation of nitrous oxide and isoflurane, and increments of fentanyl intravenously. After induction, large (14-gauge) catheters were inserted in two peripheral veins to allow acute normovolemic hemodilution. A radial artery catheter, nonheparinized central venous catheter (for sampling), and urinary catheter were also inserted. A quadruple lumen catheter (Opticath P7, 110-EH Registered Trademark; Abbott, Ottignies, Belgium) was positioned in the pulmonary artery for continuous measurement of mixed venous hemoglobin saturation (Sv O2). A bypass circuit was added to a cell saver (Dideco Autotrans Registered Trademark BT 795P; Dideco, Mirandola, Italy), so that its roller pump could be used during acute normovolemic hemodilution for accurate volume infusion. The acute normovolemic hemodilution procedure was performed as shown in Figure 1. If, 15 min after the 60 min withdrawal period, pulmonary capillary occlusion pressure (PAOP) had decreased by more than 2 mm Hg extra artificial colloid was infused in aliquots of 250 mL to restore control PAOP. Induction of anesthesia and acute normovolemic hemodilution were performed in an induction area.
All operations were performed by one of two surgeons in a standardized way according to a modified Charnley technique with the patient lying supine . Artificial colloids, packed red blood cells (PC), washed salvaged blood, and 5% albumin were transfused to keep the patient hemodynamically stable as determined by the following variables: radial systolic pressure and systolic pressure variation , Sv O2, and PAOP. After proper differential diagnosis, fluid treatment in aliquots of 250 mL was triggered by a visible increase in systolic pressure variation or a decrease in Sv O2 or PAOP by 2% or 2 mm Hg, respectively. Fluids were transfused in three subsequent phases: 1) artificial colloids alone to decrease Hct to 25%-30%; 2) artificial colloids and PC alternately (or salvaged blood if available) to keep Hct stable and to decrease total serum protein (TSP) further to 35-40 g/L; 3) 5% albumin and PC alternately until the end of operation . TSP 35-40 g/L was considered a safe concentration , allowing on average the administration of 2000 mL artificial colloid. In addition, approximate 1.5 mL centered dot kg-1 centered dot h-1 crystalloid was infused during the entire study period.
All patients were observed for 3 h. If volume expansion was necessary, the two autologous blood units, harvested during acute normovolemic hemodilution, were transfused in reversed order.
Samples and Measurements
Samples of the patients' blood were taken before (T1) and after 500 mL (T2) and 1000 mL (T3) of acute normovolemic hemodilution; intraoperatively after 20 mL/kg of artificial colloid (T4); at the end of artificial colloid infusion at [Hct] 25% and [TSP] 35-40 g/L (T5); on arrival in the recovery room (T6); and at T6+3 h (T7), Table 1. Colloid osmotic pressure and coagulation tests were performed on each sample.
Intravascular Volume End-Points
At regular intervals during acute normovolemic hemodilution, we measured heart rate, cardiac index, radial artery pressures, pulmonary artery pressures, central venous pressure, PAOP, SaO2, SvO2, PaO2, mixed venous PO2, and Hct as well as systemic and pulmonary vascular resistance, oxygen delivery, and oxygen consumption calculated according to the classical formulas.
During the entire observation period (T1 right arrow T7) all fluids necessary to maintain hemodynamic variables stable (see Intraoperative Management) were recorded.
From plasma volume measurements and records of fluids infused, we calculated extravasation of artificial colloid over the study period (Appendix 1). Plasma volume measurements were performed on the first (T1) and last sample (T7) using albumin marked with technetium-99m (t1/2 6 h) and iodine-125 (t1/2 60 days), respectively . Measurements with technetium-99m are less accurate than those with iodine-125 (coefficient of variation 5% vs 2%) because technetium-99m binds more weakly to albumin.
Plasma volume and urinary volume measurements and artificial colloid concentrations in serum and urine were used to estimate the distribution of colloid mass (Appendix 2) and volume (Appendix 3) at the end of the study. "Volume effect" was calculated as volume in serum divided by (volume in serum + extravasation). HES concentrations were determined by H. Forster (Department of Experimental Anaesthesiology, University Hospital, Frankfurt, Germany). Starch was hydrolyzed to glucose and then analyzed with hexokinase (controlled by the o-toluidine method) . To quantify GEL the colloid was hydrolyzed to 4-hydroxyproline and measured by G. De Groote (BARC Laboratories, Gent, Belgium) .
Hct and TSP were determined in the usual manner. All colloid osmotic pressures were performed by the same investigator using a Colloid Osmometer 4100 Registered Trademark (membrane cutoff 30,000 daltons; Wescor, Logan, UT).
Prothrombin time, activated partial thromboplastin time, and derived fibrinogen concentration were determined on citrated plasma using an automated coagulation analyzer ACL 810 Registered Trademark (Instrumentation Laboratory, Milan, Italy). Platelets were counted in EDTA-anticoagulated blood. Activated clotting time was measured with the Hemochron 400 Registered Trademark (International Technodyne Corporation, Metuchen, NJ) on a fresh blood sample. The same investigator performed all bleeding times using a modification of the Ivy method (Simplate II Registered Trademark; Organon Teknika, Turnhout, Belgium). The same investigator (Y.J.M.) also performed all thrombelastograms (Thromboelastograph D Registered Trademark; Hellige, Freiburg, Germany) on a fresh blood sample. Factor VIII coagulant activity and von Willebrand factor antigen were measured in a one-stage and an enzyme-linked immunosorbent assay, respectively . At both intraoperative sampling times (T4 and T5) a clinical capillary oozing score (4 = dry, 3 = moist, 2 = wet, and 1 = dripping subcutaneous tissue) was agreed upon by the surgeon and the anesthetist and the average noted. Plasma volume measurements and the recordings of all fluids infused as well as their respective Hct allowed for calculation of red cell volume loss and blood loss (Appendix 1).
Values are expressed as mean +/- SD. The homogeneity of both groups was tested (chi squared test, t-test) before statistical analysis. Repeated measurements of variables were screened for normal distribution (Kolmogorov-Smirnov test) and homogeneity of variances (Box-Cox test). Parametric variables were then further examined with two-way multivariate analysis of variance. When interaction was significant, post hoc procedures were added (Tukey highest significant difference, Scheffe). Bonferroni-modified Student's t-tests were applied to demonstrate significance at individual sampling points. One-tailed tests were used for one-sided hypotheses (e.g., blood loss in HES group >or=to blood loss in GEL group). Results significant with one-tailed testing only are indicated as such. For these results the total number of comparisons assumed in making the Bonferroni correction, and therefore applied to the one-tailed test, are also mentioned. Nonparametric data were successively analyzed by Friedman and Kruskal-Wallis tests when appropriate. Significance was accepted at P < 0.05.
All patients underwent uncomplicated primary total hip replacement with similar duration of operation and times between the standard samples. The two groups were well matched for number (n = 21), age, gender, ASA status, and height, but there was a significant difference for weight Table 2. Statistical tests were performed using weight as a covariate when appropriate. Volume of artificial colloid infused was identical (2130 +/- 350 mL): 19/21 patients in the HES group and 18/21 patients in the GEL group received 2000 mL or more of artificial colloid. No adverse drug reactions were observed.
Intravascular Volume End-Points
At the end of the acute normovolemic hemodilution procedure no patient in the HES group and only one in the GEL group required additional colloid to maintain PAOP. During transfer to the operating theatre, three additional patients in the GEL group required 250 mL artificial colloid. There were no significant differences in hemodynamic variables and oxygen balance between groups during the entire study period and values were in the normal range (SV O2, 72%-83%; PAOP, 7-15 mm Hg). Total volumes of artificial colloid (HES 2119 +/- 350 mL vs GEL 2131 +/- 516 mL), blood and colloids together (4148 +/- 1382 mL vs 3999 +/- 544 mL), crystalloid, and urine were not different Table 3. Extravasation was more pronounced in the GEL group (725 +/- 558 mL vs 382 +/- 349 mL). Serum artificial colloid concentrations were not different at any time if correction was made for the difference in concentration of the preparations (6% vs 3%) Table 1. Artificial colloid mass and volume distribution at the end of the study (T7) are shown in Table 4. At that time 76% +/- 48% of the HES volume and 56% +/- 36% of the GEL volume was still present intravascularly.
Hct decreased to the preset target (25%) in both groups Table 5. On arrival in the recovery room, Hct was significantly higher in the GEL group. In the HES group TSP decreased to the preset target (35 g/L) but remained significantly higher in the GEL group relative to the HES group after infusion of the first 500 mL of artificial colloid. Both groups showed a similar small decrease in colloid osmotic pressure Table 5.
No primary significant differences arose between groups for any coagulation test. There were four abnormal bleeding times in the HES group as opposed to none in the GEL group (P < 0.05, Fisher's exact test). Blood loss was 660 mL higher in the HES group (one-tailed t-test) Table 3. As blood loss was determined only once during the study period, no Bonferonni correction was made for this one-tailed comparison. There was poor correlation between any coagulation test and blood loss with the exception of prothrombin time (r = 0.91) . Coagulation factors were never administered before arrival in the recovery room.
Volume expansion from 500 to 1000 mL of artificial colloid infusion in the perioperative setting generally is similar to that of 4%-5% albumin [10-12]. We compared HES and GEL without reference to albumin, based upon the following three rationales. 1) Serious side effects of artificial colloid (allergy, coagulation disturbance, intra- or extravascular accumulation) are apparently related to large molecules causing the formation of immune complexes and more intense histamine release . Through their coating effect, large molecules interfere more with the function of von Willebrand factor (and, secondarily, factor VIII), with fibrin formation, and perhaps with platelet and endothelial cell function . Large molecules persist longer intravascularly and are more easily phagocytized by macrophages . 2) The prolonged intravascular volume effect of artificial colloids with a great percentage of large molecules is not needed in the perioperative period during considerable ongoing blood loss because, after a few hours, most of the colloid will have disappeared in the wound anyway . 3) In comparison to hyperoncotic preparations, isooncotic solutions have a more predictable and stable volume effect without the risk of interstitial dehydration. [3,26]. In contrast to hyperoncotic solutions, isooncotic colloids do not expand plasma volume more than the original infused volumes.
Isooncotic (3%-5%) low molecular weight dextran with MW 40,000 is not available in Belgium . In the starch group, 6% low molecular weight starch (MW 200,000) with a high substitution ratio of 0.6-0.66 (Elohaes Registered Trademark; Fresenius AG, Bad Homburg, Germany) was not selected because of its prolonged intravascular volume effect (similar to hetastarch) . The substitution ratio of starches is very important for their pharmacodynamics because the hydroxyethyl side chains markedly decelerate the intravascular breakdown by amylase . There are two subgroups of gelatins according to the manufacturing process. Urea-linked gelatin or polygelin is the result of cross-linking polypeptides with hexamethyl diisocyanate. In the other group, the degraded raw material is modified by the addition of succinic acid anhydride to become succinylated or modified fluid gelatin . A 3% modified fluid gelatin was preferred to urea-linked gelatin because of a better volume effect , less gelation tendency, and fewer allergic side effects .
If clinicians accept total protein concentrations at 25-30 g/L, albumin solutions probably could be avoided completely during otherwise uncomplicated surgery. However, at that level of protein dilution (i.e., 30%-40% of normal) the concentration of coagulation factors often will be critical, making infusion of plasma necessary . If so, plasma, which contains albumin, could be alternated with artificial colloid to maintain the protein concentration . To achieve this goal, infusion of 3000 mL or more of artificial colloid would become necessary [18,31]. The administration of such high doses has not been properly studied intraoperatively. As a first step, we diluted TSP concentration to 35-40 g/L which required infusion of approximate 2000 mL of artificial colloid .
Intravascular volume effects were studied in three different ways: 1) hemodynamic (and oxygen balance) effects after a fixed dose of artificial colloid [10,11]; 2) volume needed to achieve a defined hemodynamic end-point ; 3) direct blood volume measurement with a record of all fluids infused. Direct blood volume measurement is important for this kind of study.
Intravascular Volume End-Points
The intravascular volume effect of GEL was clearly weaker than that of HES. 1) More patients needed extra GEL after acute normovolemic hemodilution. 2) Total volumes infused were similar but a higher blood loss in the HES group counterbalanced higher extravasation in the GEL group. 3) More extravasation and a lower percent volume effect at the end of the study were recorded. 4) On arrival in recovery a higher Hct was also seen in the GEL group Table 5. A practical point should be raised here. The extravasation of GEL necessitates timely infusion of extra colloid (i.e., approximate 30% of the volume originally needed: cf. 725 mL extravasation vs 2131 mL infusion; Table 3) in order to maintain hemodilution. If not, the extravasated GEL will be replaced partially by blood according to the transfusion scheme; this will needlessly increase the Hct and blood consumption such as occurred in this study because the protocol did not allow extra infusion of GEL in between. 5) Finally, TSP was also higher in the GEL group but this might be partially due to gelatin interfering with the biuret method for protein determination .
Although weaker than HES, the perioperative intravascular volume effect of GEL seems-to be greater in this study than generally accepted [12,25]. Gelatins, due to their more pronounced extravasation (with secondarily increased interstitial pressure and lymph flow) and negative charge, could have increased the intravascular mobilization of albumin that has been described after transfusion of packed red blood cells . The similar pharmacokinetics of GEL and HES after correction for the concentration difference in the preparations are also noteworthy in this context.
No primary, between-group differences were found for any coagulation test. This relative stability of coagulation tests after HES has been described in other studies . When abnormal coagulation tests were considered separately, more abnormal bleeding times were found in the HES group. Putting aside a possible direct effect of HES on platelet aggregation , this increase in abnormal bleeding times could reflect interference with the release (or maybe even assembly) of von Willebrand factor and, secondarily, with factor VIII .
Average blood loss was high in both groups. The blood loss measured in this study, was approximately 1000 mL greater than that estimated by our anesthetists in daily practice. Anesthetists seem to underestimate higher blood losses, especially the part adsorbed in the surgical drapes. A second reason is related to the surgical technique . Intraoperatively, a silicone mold was made from the femur shaft. A computer reshaped a standard prosthesis to match the model. This process took 30-45 min while the patient was waiting with an open femur.
A higher blood loss was found in the HES group through one-tailed testing only: gelatins have never been reported to cause more blood loss than any other kind of colloid and they do not influence coagulation other than by dilution. The relatively weak significance could be partially related to the small power (65% for 600 mL difference). Increased blood loss with HES was not found in other studies (). However, blood volume was not measured in these publications and HES was compared to albumin. Albumin might impair coagulation .
This study quantified a weaker intravascular volume effect of GEL and a higher blood loss with HES after 2000 mL of artificial colloid infusion. The higher blood loss was significant with one-tailed testing only. Two practical conclusions can be drawn. First, during hemodilution extra GEL (i.e., approximate 30% of the volume originally needed) must be administered to avoid secondary hemoconcentration as the GEL extravasates. Second, infusion of 2000 mL HES requires careful attention to measurements of coagulation.
Blood Loss and Extravasation
Red cell volume (RCV) loss = RCVT1 - RCVT7 + RCV infused
Blood loss (BL) = RCV loss divided by global average Hct
(Global average Hct = average of average Hcts between two subsequent samples weighted according to the volume infused between these samples.)
Total volume infused (TV) = artificial colloid (AC) + PC + A 5% + salvaged blood
Extravasation of colloid = (TV - BL) + (BVT1 - BVT7)
Distribution of AC Mass at End of Observation
Mass AC in serum = serum AC concentrationT7 times plasma volume (T7)
Mass AC in urine = sum of urinary AC concentration at T5, T6, and T7, each multiplied with its respective urinary volume
Mass AC in blood loss = AC concentration in second unit acute normovolumic hemodilution times 500 + average blood AC concentration between T3 and T6 times the infused volume over that period + analogous for the period T6 to T7. (Correction was made for the difference in extravasation between HES and GEL.)
Mass AC extravasated = total mass AC administered - the three previous masses
Distribution of AC Volume at End of Observation
Volume AC in serum = (total volume AC infused - extravasation) times mass in serum divide by (mass in serum + mass in BL)
Volume AC in urine = 0 mL (urine output is assumed to be maintained by crystalloid infusion).
Volume AC in BL = (total volume AC infused - extravasation) times mass in blood loss divide by (mass in serum + mass in blood loss).
Volume AC extravasated = extravasation (see Appendix 1)
1. The Belgium SAnGUIS Study Group: Baele PL, De Bruyere M, Deneys V, et al. Results of the SAnGUIS study in Belgium. A concerted action of the Commission of the European Communities. IVth Medical and health Research Programme. Acta Chir Belg 1994;94 Suppl:5-61.
2. Adams HA, Michels R, Hempelmann G. Indications for the use of human albumin in anesthesia and intensive care medicine. Anesthesiol Intensivmed Notfallmed Schmerzther 1993;28:66-74.
3. Kohler H, Zschiedrich H, Linfante A, et al. Elimination of hydroxyethyl starch 200/0.5, dextran 40, and oxypolygelatin. Klin Wochenschr 1982;60:293-301.
4. Strauss RG, Stanfield C, Heinriksen RA, Villhauer PJ. Pentastarch may cause fewer effects on coagulation than hetastarch. Transfusion 1988;28:257-60.
5. Kroll W, Gerner P, Polz W. Vergleichende Untersuchung zur Volumenwirkung korperfremde Kolloide. Perfusion 1993;7:286-99.
6. Lundsgaard-Hansen P, Tschirren B. Clinical experience with 120,000 units of modified fluid gelatin. Dev Biol Stand 1980;48:251-6.
7. Laxenaire M-C, Charpentier C, Feldmann L et le Groupe Francais d'Etude de la Tolerance des Substituts Plasmatiques. Anaphylactoid reactions to colloid plasma substitutes: frequency, risk factors, mechanisms. A French prospective multicentre enquiry. Ann Fr Anesth Reanim 1994;13:301-10.
8. Hogan JJ. The use of colloidal (gelatin) solutions in shock. JAMA 1915;64:721-6.
9. Van der Linden Ph, Schmartz D. Pharmacology of gelatins. In: Baron J-F, ed. Plasma volume expansion. Paris: Arnette Blackwell, 1992:67-73.
10. Boldt J, Kling D, Zickmann B, et al. The hemodynamic effects of various hydroxyethyl starch solutions in cardiosurgery patients. Anaesthesist 1990;39:6-12.
11. Hankeln K, Senker R, Beez M. Comparative study of the intraoperative efficacy of 5% human albumin and 10% hydroxyethyl starch (HAES-steril Registered Trademark) in terms of hemodynamics and oxygen transport in 40 patients. Infusionstherapie 1990;17:135-40.
12. Edwards JD, Nightingale P, Wilkins RG, Faragher EB. Hemodynamic and oxygen transport response to modified fluid gelatin in critically ill patients. Crit Care Med 1989;17:996-8.
13. Waldhausen P, Kiesewetter H, Leipnitz G, et al. Hydroxyethyl starch-induced transient renal failure in preexisting glomerular damage. Acta Med Austriaca 1991;18(Suppl 1):52-5.
14. Tonnessen T, Tollofsrud S, Kongsgaard UE, Noddeland H. Colloid osmotic pressure of plasma replacement fluids. Acta Anaesthesiol Scand 1993;37:424-6.
15. Mulier JC, Mulier M, Bradly LP, et al. A new system to produce intraoperatively custom femoral prosthesis from measurements taken during the surgical procedure. Clin Orthop 1989;249:97-112.
16. Pizov R, Segaz E, Kaplan L, Perel A. The use of systolic pressure variation in hemodynamic monitoring during deliberate hypotension in spine surgery. J Clin Anesth 1990;2:96-100.
17. Kamal GD, Tatman BA, Symreng T. Mixed venous blood is a more sensitive indicator of acute blood loss than arterial blood in swine. Anesth Analg 1993;76:S176.
18. Mortelmans Y, Vermaut G, Van Aken H. A simple method for calculating component dilution during fluid resuscitation (the Leuven approach). J Clin Anesth 1994;6:279-87.
19. Kowalsky JR, Perry JR. Radiopharmaceuticals in nuclear medicine practice. 1st ed. Norwalk: Appleton & Lange, 1987:411-20.
20. Forster H, Wicarkzyk C, Dudziak R. Bestimmung der Plasmaelimination von Hydroxyaethylstarke und von Dextran mittels verbesserter analytischer Methodik. Infusionstherapie 1981;2:88-94.
21. Dawson CD, Jewell S, Driskell J. Liquid-chromatographic determination of total hydroxyproline in urine. Clin Chem 1988;34:1572-4.
22. Tornay I, Declerck PJ, Smets L, et al. Measurement of von Willebrand factor antigen in plasma and platelets with an enzyme-linked immunosorbent assay based on two murine monoclonal antibodies. Haemostasis 1991;21:125-34.
23. Ring J. Anaphylactoid reactions to intravenous solutions used for volume substitution. Clin Rev Allergy 1991;9:397-414.
24. Strauss RG. Volume replacement and coagulation: a comparative review. J Cardiothorac Anesth 1988;2(Suppl 1):24-32.
25. Quon CY. Clinical pharmacokinetics and pharmacodynamics of colloidal plasma volume expanders. J Cardiothorac Anesth 1988;2(Suppl 1):13-23.
26. Kohler H, Zschiedrich H, Clasen R, et al. Blood volume, colloid osmotic pressure and renal function in volunteers after infusion of middle-molecular 10% hydroxyethyl starch 200/0.5 and 10% dextran 40. Anaesthesist 1982;31:61-7.
27. George M, Griffith N, Lehot JJ, et al. Comparaison dextran 40 a 3.5% versus albumine a 4% utilisees comme solutes d'expansion volemique apres chirurgie cardiaque. Effets sur les parametres hemodynamiques et la pression oncotique colloidale. Cah Anesthesiol 1993;41:18-22.
28. Mortelmans Y, Merckx E, Van Nerom C, et al. Effect of an equal volume replacement with 500 mL 6% hydroxyethyl starch on the blood and plasma volume of healthy volunteers. Eur J Anaesthesiol 1995;12:259-64.
29. Himpe D, Van Cauwelaert P, Neels H, et al. Priming solutions for cardiopulmonary bypass: comparison of three colloids. J Cardiothorac Vasc Anesth 1991;5:457-66.
30. NIH Consensus Conference. Fresh-frozen plasma: indications and risks. JAMA 1985;253:551-3.
31. Sommoggy St, Fraunhofer J, Jelen-Esselborn S, Stemberger A. Coagulation changes during aortofemoral bifurcation bypass operations: is plasma substitution possible with hydroxyethyl starch (HES) only? Anaesthesist 1990;39:353-60.
32. Boldt J, Knothe C, Zickmann B, et al. Cardiorespiratory and microcirculatory effects following volume replacement using a new hydroxyethyl starch preparation. Anaesthesist 1992;41:316-23.
33. Doumas BT, Bayse DD. A candidate reference method for determination of total protein in serum. I. Development and validation. II. Test for transferability. Clin Chem 1981;27:1642-51.
34. Altschule MD. Vagaries of transcapillary albumin flux. Chest 1986;90:280-1.
35. London MJ, Ho JS, Triedman JK, et al. A randomized clinical trial of 10% pentastarch (low molecular weight hydroxyethylstarch) versus 5% albumin for plasma volume expansion after cardiac operations. J Thorac Cardiovasc Surg 1989;97:785-97.
36. Boldt J, Knothe J, Zickmann B, et al. Influence of different intravascular volume therapies on platelet function in patients undergoing cardiopulmonary bypass. Anesth Analg 1993;76:1185-90.
37. Vogt N, Bothner U, Georgieff M. Comparison of human albumin 5% and 6% HES 200/0.5 as exclusive colloid component in major surgery. Anasthesiol Intensivmed Notfallmed Schmerzther 1994;29:150-6.
© 1995 International Anesthesia Research Society
38. Dietrich G, Orth D, Haupt W, Kretschmer V. Primary hemostasis in hemodilution. 2.) Infusion solutions. Infusionsther 1990;17:214-6.