• Fluid balance; albumin;
  • hydroxyethyl;
  • starch


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

We studied the long-term efficacy and safety of medium-molecular-weight hydroxyethyl starch (HES) administered in doses above 20 mlkg−1 during major blood replacement therapy. Blood replacement for 50 patients used 6% HES 200/0.5 (HES group) or 5% albumin (ALB group) and additional blood components according to a defined protocol. We compared safety, efficacy and costs in 4 peri-operative days. Colloid administration on the day of surgery was 38.4 mlkg−1 (HES group) and 35.1 mlkg−1 (ALB group). Haemodynamic, coagulation and renal function parameters were similar. Although total serum protein was still different on the third postoperative day (53.45 gl−1 (HES group) and 60.6 gl−1 (ALB group) (p < 0.01)) the colloid osmotic pressure always remained above 19.5 (2.5) mmHg (HES group). Blood loss (3810 (1632) ml (HES group) and 3455 (1733) ml (ALB group)) and the requirement for blood components was comparable. Costs were reduced by 35% (p < 0.05) in the HES group. We conclude that using 6% HES 200/0.5 as the only colloid for treatment even of large blood loss is a safe and economic alternative to albumin.

Blood replacement therapy has contributed to the success of surgical procedures like major cancer operations or trauma surgery. Blood components present problems owing to the risks inherent in blood transfusions and to the high cost of blood preparations. Future strategy must involve reducing the amount of blood used and increasing the safety of blood components [1]. This includes autologous transfusion [2] or surgical methods designed to reduce blood loss. The use of blood components should also be limited to the minimum necessary to maintain blood function. Different requirements exist for oxygen transport, oncotic and haemostatic functions, and as the circulating blood volume must always be fully maintained, particularly during haemodilution, plasma expanders are of major importance in blood replacement therapy. For a colloid requirement above 1.5–2−1 natural colloids in the form of human albumin or plasma protein solutions are commonly used, but this remains costly [3]. Use of albumin is justified by the need to avoid hypoproteinemia, and to preserve oncotic plasma function or by the dose limits of artificial colloids. The dose limits of artificial colloids are based on specific interactions with clotting. In terms of clotting interaction, solutions of high-molecular-weight, highly substituted HES have virtually the same side-effects as dextran, whereas solutions of medium-molecular-weight HES have less effect on clotting [4, 5] and may have an advantage over gelatine solutions in the incidence of anaphylactic reactions [6]. We do not know the long-term effects of high-dose HES administration (>20−1) over several days, or whether 6% HES 200/0.5 as a medium-molecular-weight HES, can replace 5% human albumin solution for the entire peri-operative period, even when used in amounts exceeding 20 ml. kg−−1.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

The study was approved by the Ethics Committee of the University of Ulm. After obtaining patients' informed written consent, we selected 50 male urological patients who had to undergo radical prostatectomy or cystectomy with bladder replacement (Table 1). Patients were not studied if: weight <60 kg, age <21 years, ASA 1 or 2, haemoglobin <12 g.l−1, history of clotting disorders (prothrombin time <75%, activated partial thromboplastin time (aPTT) >45 s, platelet count <100 g.l−1), liver function disorders (elevated transaminases), advanced renal insufficiency (creatinine >250 mmol.l−1) or hypoproteinemia (total serum protein <45 g.l−1).

Table 1.  Demographic characterisation of study groups Thumbnail image of
  • a

    HES = hydroxyethyl starch, ALB = albumin. Data are mean (SD).

  • During the peri-operative observation period, measurements were made pre-operatively, at the end of surgery and on the morning of the 1st and 3rd postoperative days. The haemodynamic parameters were heart rate, blood pressure and central venous pressure (CVP). Blood samples were taken for: haemoglobin, leucocyte and platelet counts, total serum protein (TSP), colloid osmotic pressure (COP) with the oncometer (membrane exclusion limit 20 000 Da), prothrombin time, aPTT, glucose, serum lactate (reference: 0.5–2.2 mmol.l−1), electrolytes, creatinine, urea and serum osmolality. We also performed blood gas analysis and pulse oximetry. Postoperatively, we determined the total blood loss from the measured (suction canister) and estimated blood loss (weighed towels and swabs), recorded the blood loss from wound drainage and continuously measured the urine volume. At the end of the study, we determined the intake and output fluid balance. From this, on the basis of current average prices, we calculated the cost of fluid replacement therapy (6% HES 200/0.5: (£7.50/500 ml), 5% human albumin: (£45.00/400 ml), packed red blood cells (PRBC): (£56.00/unit), fresh frozen plasma (FFP): (£45.00/unit)). Moreover, we recorded any reactions and postoperative bleeding requiring intervention.


    All patients underwent surgery using standard general anaesthetic. Induction was performed with fentanyl 0.01 mg. kg−1, midazolam 0.1−1 and pancuronium 0.1 mg. kg−1. Maintenance was with fentanyl, midazolam and, when necessary, enflurane 0.4–0.6 vol.%. Muscle relaxation was maintained with pancuronium under the control of a nerve stimulator. Controlled ventilation was achieved with a nitrous oxide/oxygen mixture (Fio2 0.35 and a PE′co2 of 36–38 mmHg). The patients were transferred after operation to an intensive care unit and their lungs were ventilated. The trachea was extubated on the evening of the day of surgery.

    Depending on blood volume requirements (clinical estimation based on blood losses and haemodynamic monitoring parameters), intra-operative blood replacement was performed in a standardised manner according to a modified Bernese component scheme [7]. For a volume requirement of up to 1000 ml, we administered a colloidal solution, for a requirement from 1000 to 4500 ml a colloidal solution and packed red blood cells in a 1:1 ratio and above 4500 ml packed red blood cells and FFP in a 1:1 ratio. In addition, electrolyte solutions were infused at a rate of 8−1.h−1.

    Postoperative therapy with blood components was controlled on the basis of specific intervention thresholds depending on the blood composition. Colloidal solutions were administered when circulating volume was reduced as determined by haemodynamic changes or clinical evaluation. For packed red blood cells transfusion, we set a haemoglobin limit of 11 g.dl−1 and for FFP, a prothrombin time limit of 50% or a 50-s limit for aPTT. When we suspected a clotting disorder due to platelet deficiency, platelets would be administered when the platelet concentration is <50 g.l−1. As basic fluid replacement therapy, the patients received a daily infusion of crystalloid solutions at an infusion rate of 40−1. Dopamine 2 μg. kg−1.min−1 was administered continuously to all patients to optimise perfusion and to increase diuresis, and heparin was given at 15 000 IU.24 h−1 to prevent thrombosis.


    The patients were assigned to the ALB or the HES group using random numbers. The ALB group received 5% human albumin and the HES group 6% hydroxyethyl starch with an average molecular weight of 200 000 Da and a degree of substitution (DS) of 0.5 (6% HES 200/0.5), as the only colloid.


    The variables are presented as mean (SD). After confirming normal distribution, we tested group differences and time influences using two-way repeated measures anova with multiple testing using Student–Newman–Keuls test at a significance level of p < 0.05 and p < 0.01. For demographic data, blood balances and costs we used Student's t-test (p < 0.05). The ASA status and the number of units of administered PRBC, FFP and platelets were tested by the Chi-squared test (p < 0.05).


    1. Top of page
    2. Abstract
    3. Methods
    4. Results
    5. Discussion
    6. References

    The groups were comparable in terms of demographic data, pre-existing diseases and ASA risk classification (Table 1).

    Haemodynamic functions were within normal physiological limits. A difference in CVP was noted only at the end of surgery, being 1.4 mmHg lower in the ALB group. Overall, stable haemodynamic conditions were found at all points of measurement (Table 2).

    Table 2.  Haemodynamic parameters, blood gas analysis, and metabolism Thumbnail image of
  • a

    HR = heart rate, MAP = mean arterial pressure, CVP = central venous pressure, HES = hydroxyethyl starch, ALB = albumin. Data are mean (SD). * p < 0.05 (influence of time); † p < 0.05 (between groups).

  • The variations of concomitant metabolic parameters, such as the glucose level, were mostly in the same direction and during the peri-operative stress phase showed an increase to slightly elevated values. Serum lactate levels were comparable and metabolic acidosis was absent in both groups (Table 2).

    At the end of surgery, Pao2 and saturation were elevated during mechanical ventilation at Fio2= 0.35, but by the end of the study had returned to baseline values breathing room air. No group differences were found (Table 2).

    The haemoglobin level reduced at first in both groups, but then remained within a range of 10–11.5 g.dl−1. Group differences were found on the first postoperative day, but at the end of the study concentrations were again comparable. The platelet counts decreased initially and did not show any group differences (Table 3).

    Table 3.  Haematology, haemostasis, and oncotic function Thumbnail image of
  • a

    aPTT = activated partial thromboplastin time, TSP = total serum protein, COP = colloid osmotic pressure. HES = hydroxyethyl starch, ALB = albumin. Data are mean (SD). * p < 0.05 (influence of time); † p < 0.01 (between groups).

  • Total serum protein (TSP) levels showed clear differences. In the ALB group, TSP reduced from 66 g.l−1 to a minimum level of 56 g.l−1, whereas the patients in the HES group had a mean TSP level of only 37 g.l−1 immediately postoperatively, which rose to 53 g.l−1 at the end of the study period. In the HES group, on the first postoperative day, the COP dropped to 18.5 mmHg vs. 22.3 mmHg (ALB), then increased to 19.5 mmHg, but during the entire study always remained significantly lower than in the albumin group (23.6 mmHg) (Table 3).

    The prothrombin time decreased during the peri-operative period in both groups, lowest immediately following surgery and increasing to values of more than 80% by the end of the study period. APTT at the end of surgery, however, was significantly longer in the HES group than in the ALB group and from this time onward both the HES and the ALB groups showed slightly prolonged but comparable values up to the third postoperative day (Table 3). Renal function creatinine, urea and diuresis was the same in both groups (Table 4).

    Table 4.  Osmolality, renal function and water balance Thumbnail image of
  • a

    RL = Ringers lactate. HES = hydroxyethyl starch, ALB = albumin. Data are mean (SD). * p < 0.05 (influence of time).

  • Neither the number who received blood components nor the frequency distribution of blood components over the two groups showed any differences (Table 5). The amounts of colloid, PRBC and FFP used for replacement were comparable in all cases. No platelets were transfused (Table 6).

    Table 5.  Distribution of units of blood components between the groups Thumbnail image of
  • a

    PRBC = packed red blood cells, FFP = fresh frozen plasma, HES = hydroxyethyl starch, ALB = albumin.

  • Table 6.  Balance and costs of haemotherapy Thumbnail image of
  • a

    PRBC = packed red blood cells, FFP = fresh frozen plasma. HES = hydroxyethyl starch, ALB = albumin. Data are mean (SD). * p < 0.05.

  • Most of the blood components were administered during surgery. Thus, patients in the HES group received an average 2.89 l (38.4 (11)−1) and those in the ALB group 2.6 l (35.1 (10)−1) of colloids on the day of surgery (Table 6).

    At the end of the study, the average cost per patient was £482 in the HES group and £750 in the ALB group (p < 0.05) (Table 6). The two groups showed a comparable total blood loss both during surgery and in the postoperative drainage (Table 6).


    1. Top of page
    2. Abstract
    3. Methods
    4. Results
    5. Discussion
    6. References

    Because of its high cost and limited availability, various attempts are being made to reduce albumin administration [8]. To study the suitability of large-dose artificial colloids as an alterative requires a comparison of the efficacy and side-effects of colloid preparations.

    The efficacy of colloidal plasma expanders depends on their intravascular concentration, molecular size and pharmacokinetic properties, i.e. the magnitude and the duration of the volume effect, and oncotic stabilisation, determined by haemodynamic variables but also by organ function and metabolism. The evaluation of the colloidal volume effect varies with the circumstances under which it is infused. While clinical efficacy can be demonstrated with a multitude of crystalloid solutions [9], a moderate volume effect that lasts several hours is advantageous in that it provides volume stability even under marked haemodilution and therefore maintains O2 transport capacity.

    In this study, we found that the haemodynamic effect of HES was comparable to that of 5% albumin throughout the whole period, and that haemodynamic-related variables remained within physiological limits. In view of the short half-lives (<4 h) of solutions of medium-molecular-weight, medium-substituted HES, we could not have predicted that comparable haemodynamic stability would be maintained for 3 days with similar amounts of colloids.

    Oncotic function was stable in the HES group. As expected after major abdominal surgery [10], the TSP level in patients of the HES group was low. It was less than 40 g.l−1 at the end of surgery and far below the TSP limit of 45–50 g.l−1 postulated by some investigators [11, 12] in order to prevent the risk of hypoproteinaemic fluid changes. The colloid osmotic pressures measured, however, were always within a safe reference range even though they did not quite match the level in the ALB group. Kinetic properties of 6% HES 200/0.5 are such that this solution, when administered in an amount comparable to that of 5% human albumin, can close quite a considerable oncotic gap brought about by a temporary but large blood loss, until endogenous colloidal substances again become available.


    A volume expander is to be viewed as ideal, if it is biologically inert and completely eliminated as soon as endogenous colloids ensure oncotic function. No artificial colloid currently meets such requirements in terms of side-effects [13]. As a result, limits have been recommended to their administration. This is due primarily to complex haemostatic interactions with endogenous plasma coagulation, fibrinolysis or platelet function, which for dextran solutions exceeded those effects caused by dilution and made it desirable to limit the daily dose to 20−1 [14].

    Because early solutions of HES with an average molecular weight of 450 000 Da and a DS of 0.7 caused similar clotting disturbances to dextran, this recommendation was first applied to all HES preparations. With newer HES solutions, it became clear that smaller molecules and reduced DS affected not only the elimination kinetics but also the side-effects [4, 15].

    A large number of reports based on controlled studies or case reports on interactions with clotting reflect a striking heterogeneity and made a consistent evaluation of the haemostatic characteristics of HES solutions difficult [4, 16[17][18]–19].

    Effects on endogenous coagulation, especially factor VIII components [20], must be regarded as a confirmed interaction of HES solutions and clotting functions. We found a moderate postoperative increase in aPTT in the HES group compared with the ALB group which could be interpreted as a factor VIII interaction. In the subsequence period, however, this increase was abolished by the low-dose heparinization routinely used from the first postoperative day onwards and which altered aPTT of both groups. We found no indication of potentiated heparin action after a large HES dose. No recognisable clinical effects on haemostatic function were observed on the basis of either intra-operative blood loss, the wound drainage loss or requirements for clotting-promoting plasma. Even for an average daily dose of 38.4−1 of HES solution on the day of surgery, we observed no relevant HES-induced haemostatic disorders during the subsequent postoperative period. Besides, a relationship between HES administration and impairment of renal function has been discussed in recent publications [21]. Because morphological changes of tubular cells in donor kidneys were similar to changes observed in osmotic nephrosis, the potential nephrotoxicity of HES solutions could not be excluded. On the other hand, despite widespread HES use even as colloidal additive to preservative solutions for transplant organs [22], there are only a few case histories of HES-related renal impairment [23]. Such impairment can attain significance presumably only when several risk factors coincide, such as pre-existing function impairment, long-term hyperoncotic HES administration for dilutional treatment or marked dehydration with inadequate monitoring of the water–electrolyte balance or of renal function. In addition, indications exist that chemically different HES solutions can have different effects on renal function parameters [24]. Hence, it appears necessary to limit statements concerning renal HES interactions to the chemically defined HES preparation studied. Although we did not exclude patients with incipient renal insufficiency from the study, serum homeostasis was maintained and we found no indication on the basis of serum creatinine, serum urea or diuresis that 6% HES 200/0.5 compared with 5% human albumin caused impairment of renal function, provided crystalloid administration was adequate and the administration of hyperoncotic HES solutions was omitted.

    Intolerance reactions were not noted. Although they could not be excluded, they were not expected in this study because of the small size of the groups and the low incidence described previously [6].

    The cost of therapy in the HES group was 35.2% lower than in the ALB group. In terms of overall health-care costs, the use of artificial colloids thus provides an economical, low-risk blood replacement alternative especially in procedures involving pronounced bleeding.

    We can conclude that a defined 6% solution of medium-molecular-weight, medium-substituted HES 200/0.5 is an economical alternative to human colloids even in cases of large blood turnover and that, with adequate monitoring of vital and homeostatic functions and observance of contraindications, it can safely replace these.


    1. Top of page
    2. Abstract
    3. Methods
    4. Results
    5. Discussion
    6. References
    • 1
      Irving GA. Perioperative blood and blood component therapy. Canadian Journal of Anaesthesiology 1992; 39: 110515.
    • 2
      Chernow B. Blood conservation in critical care — The evidence accumulates. Critical Care Medicine 1993; 21: 4812.
    • 3
      McClelland DB. ABC of tranfusion. Human albumin solutions. British Medical Journal 1990; 300: 357.
    • 4
      Strauss RG, Stansfield C, Henriksen RA, Villhauer PJ. Pentastarch may cause fewer effects on coagulation than hetastarch. Transfusion 1988; 28: 25760.
    • 5
      Vogt NH, Bothner U, Lerch G, Lindner KH, Georgieff M. Large-dose administration of 6% hydroxyethyl starch 200/0.5 for total hip arthroplasty: plasma homeostasis, haemostasis, and renal function compared to use of 5% human albumin. Anesthesia and Analgesia 1996; 83: 2628.
    • 6
      Laxenaire MC, Charpentier C, Feldman L. Anaphylactoid reactions to colloid plasma substitutes: incidence, risk factors, mechanisms. A French multicenter prospective study. Annales Francaises d'Anesthesie et de Reanimation 1994; 13: 30110.
    • 7
      Lundsgaard-Hansen P & Pappova E. Component cell therapy of surgical haemorrhage: red cells, plasma substitutes and albumin. Annals of Clinical Research 1981; 13 2: 2638.
    • 8
      Durand Zaleski I, Bonnet F, Rochant H, Bierling P, Lemaire F. Usefulness of consensus conferences: the case of albumin. Lancet 1992; 340: 138890.
    • 9
      Moss GS, Lowe RJ, Jilek J, Levine HD. Colloid or crystalloid in the resuscitation of haemorrhagic shock: a controlled clinical trial. Surgery 1981; 89: 4348.
    • 10
      Sun X, Iles M, Weissman C. Physiologic variables and fluid resuscitation in the postoperative intensive care unit patient. Critical Care Medicine 1993; 21: 55561.
    • 11
      Kaminski M & Williams S. Review of the rapid normalization of serum albumin with modified tatal parenteral nutrition solutions. Critical Care Medicine 1990; 18: 327.
    • 12
      Lundsgaard-Hansen P. Physiology and pathophysiology of colloid osmotic pressure and albumin metabolism. Current Studies of Hematology and Blood Transfusion 1986; 117.
    • 13
      Nearman HS & Herman ML. Toxic effects of colloids in the intensive care unit. Critical Care Clinics 1991; 7: 71323.
    • 14
      Bergqvist D. Dextran and haemostasis. A review. Acta Chirurgica Scandinavica 1982; 148: 63340.
    • 15
      London MJ, Franks M, Verrier ED, Merrick SH, Levin J, Mangano DT. The safety and efficacy of ten percent pentastarch as a cardiopulmonary bypass priming solution. A randomized clinical trial. Journal of Thoracic and Cardiovascular Surgery 1992; 104: 28496.
    • 16
      Shatney CH, Deepika K, Militello PR, Majerus TC, Dawson RB. Efficacy of hetastarch in the resuscitation of patients with multisystem trauma and shock. Archives of Surgery 1983; 118: 8049.
    • 17
      Villarino ME, Gordon SM, Valdon C, et al A cluster of severe postoperative bleeding following open heart surgery. Infection Control and Hospital Epidemiology 1992; 13: 2827.
    • 18
      Dalrymple-Hay M, Aitchison R, Collins P, Sekhar M, Colvin B. Hydroxyethyl starch induced acquired von Willdebrand's disease. Clinical and Laboratory Haematology 1992; 14: 20911.
    • 19
      London MJ, Ho JS, Triedman JK, et al A randomized clinical trial of 10% pentastarch (low molecular weight hydroxyethyl starch) versus 5% albumin for plasma volume expansion after cardiac operations. Journal of Thoracic and Cardiovascular Surgery 1989; 97: 78597.
    • 20
      Kapiotis S, Quehenberger P, Eichler H-H, et al Effect of hydroxyethyl starch on the activity of blood coagulation and fibrinolysis in healthy volunteers: comparison with albumin. Critical Care Medicine 1994; 22: 60612.
    • 21
      Legendre C, Thervet E, Page B, Percheron A, Noël LH, Kreis H. Hydroxyethylstarch and osmotic-nephrosis-like lesions in kidney transplantation. Lancet 1993; 342: 2489.
    • 22
      Schlumpf R, Morel P, Loveras JJ, et al Examination of the role of the colloids hydroxyethylstarch, dextran, human albumin, and plasma proteins in a modified UW solution. Transplantation Proceedings 1991; 23: 23625.
    • 23
      Waldhausen P, Kiesewetter H, Leipnitz G, et al Hydroxyethyl starch-induced transient renal failure in preexisting glomerular damage. Acta Medica Austriaca 1991; 18 2: 525.
    • 24
      Biguzas M, Jablonski P, Thomas AC, et al Evaluation of UW solution in a rat kidney preservation model. I. Effect of hydroxyethyl start and electrolyte composition. Transplantation 1990; 49: 8725.