Intravenous infusion of crystalloid and colloid fluids is a therapeutic action intrinsic to surgical and intensive care. The volume of fluid is a controversial issue. Guidance has often come from studies wherein exogenous tracers show that plasma volume expansion in response to crystalloid fluid (‘fluid efficacy’) is poor. These results are at odds with experiences from intensive care where the requirements for crystalloid and colloid fluids appear to be quite similar.
Results obtained by tracer methods during the past 50 years are very consistent. Using the ratio of Evans blue dye to bromide dilution, Moore et al.1 found that only 22% of the Ringer's lactate infused after blood withdrawal in volunteers remained in the plasma. Lamke and Liljedahl2 using iodine-labelled albumin reported that the plasma volume increased by 17% of the volume of isotonic saline infused over 90 min in the postoperative period. Koomans et al.3 showed that 18% of infused saline expanded the plasma volume 45 to 60 min after administration ended. Recently, Jacob et al.4 reported that the intravascular volume effect of Ringer's lactate amounts to 17%.
These results agree with the relative sizes of the plasma versus interstitial fluid space volumes.4 Medical textbooks, therefore, typically state that crystalloid fluid expands the blood volume by 20% of the infused amount (‘fluid efficacy’). Recently, intensive care journals recommended the replacement of five times the amount of haemorrhaged blood volume with Ringer's solution.4,5
Clinical studies tell quite a different story to the one obtained by tracer studies. Given that hydroxyethyl starch (HES) is designed to expand the blood volume by the infused amount, one would expect that five times more crystalloid fluid than HES would be needed to reach the same clinical endpoint (urinary excretion, arterial pressure or stroke volume). No study shows this to be the case. Titration of fluid during colorectal surgery resulted in only 15% more Ringer's lactate than HES to maintain stroke volume.6 In the 6S trial, the amounts of infused Ringer's acetate and HES were virtually identical,7 and only 25 to 32% more crystalloid than HES was required in the well known VISEP and CHEST studies.
A study in which 900 ml of blood was withdrawn from volunteers showed that an infusion of 900 ml of Ringer's acetate completely restored the pulmonary artery and central venous pressures, whereas cardiac output remained 4% below baseline.8 A further 900 ml of Ringer's acetate overcompensated all meaningful volume, flow and pressure endpoints. Here, the fluid efficacy must clearly have been between 50 and 100%.
Effects of nonsteady state
The differences in fluid efficacy suggested by tracer and physiological end-point studies raise the question of whether one or both of these methodologies is erroneous. My answer is that both methodologies are likely to be correct. To understand why this is the case, we have to appreciate that the efficacy of crystalloid fluid is greatly time-dependent and increased by three factors: distribution; hypotension; and surgery-induced prolongation of the half-life. The impact of these factors will be discussed below.
Transfer of excess fluid across the capillary membrane occurs very quickly. For example, in the hand, equilibration is reached within 2 min. Equilibration of infused crystalloid fluid in the whole extracellular fluid space, however, requires much a greater period of time.
Kinetic studies performed during the past 15 years have repeatedly shown that crystalloids distribute with a half-time of about 8 min, which means that equilibration requires 30 min to be complete.9 Several nonkinetic studies also support this finding, although its clinical consequence has rarely been acknowledged.3
This slow distribution increases the efficacy of crystalloid fluid to a greater degree than that indicated by tracer methods, for the duration of the infusion and for up to 30 min thereafter. In one study, the fluid efficacy averaged 60% for Ringer[Combining Acute Accent]s acetate during 30 transurethral prostatic resections performed under general anaesthesia.10 This figure is likely to be what the anaesthetist experiences during routine surgery.
In 1965 Henry Swan11 described that a decrease in arterial pressure slowly increased plasma volume in dogs. This effect is expected on the basis of the Starling equation and is immediate if hypotension occurs during an ongoing infusion. Hypotension caused by spinal or epidural anaesthesia increases fluid efficacy from 30 to 60%.12 From a kinetic point of view, a decrease in the mean arterial pressure by 20 to 30% arrests the distribution of infused fluid for 20 to 30 min, probably depending on the infusion rate. During this time, the fluid efficacy is 100%. A new equilibrium will finally be reached but at a relatively larger plasma volume.9
Many physiological alterations increase the fluid efficacy by acting on urinary excretion. Marked fluid retention occurs during laparoscopic surgery performed under general anaesthesia with positive-pressure ventilation. Here, the diuretic response is only 5 to 10% of that seen in conscious volunteers.9,13 Mild dehydration, nonhypotensive hypovolaemia and isoflurane anaesthesia with spontaneous respiration, all reduce diuretic response by 50% or more.
The impact of the poor diuretic response to crystalloid fluid is that urinary excretion becomes a useless index of hypervolaemia. Moreover, the efficacies of commonly used crystalloid and colloid solutions will gradually approach each other. This is illustrated by the kinetic simulation described below (Fig. 1).
A hypothetical situation was created that compared plasma volume expansion following infusion of 500 ml of two widely used colloid fluids (albumin 5% and HES 130/0.4).14,15 The same amount of Ringer's acetate was also infused, using one set of kinetic variables valid for conscious volunteers, and another for general anaesthesia and surgery.
As expected, fluid efficacies were similar for all fluids during short infusions (Fig. 1a). The average efficacy of Ringer's acetate during a 10-min infusion was 80% (Fig. 1c) and is unlikely to be different if the infused volume is changed (within reasonable limits). Marked differences in fluid efficacy between the crystalloids and colloids did not occur until the infusion was turned off.
Infusing 4000 ml of these fluids over 24 h is a possible situation in the ICU. Ultimately, plasma volume expansion from colloids will be similar to that obtained by crystalloid (Fig. 1b). Fluid efficacy is about 10% in both cases (Fig. 1d). These calculations assume that the diuretic response to crystalloid fluid in the ICU is reduced in a similar way to the reduction that occurs during anaesthesia and surgery. The same result is obtained if kinetic simulations are performed with and without modelling of potentially uneven distributions of infused fluid between the plasma and interstitial fluid space.13
Unfortunately, we have no data on the volume kinetics of colloid fluid during anaesthesia, surgery or ICU care. In volunteers, the duration of the plasma volume expansion after infusion of albumin 5% is governed by the capillary leak,14 which is typically increased in severe disease. If anything, the fluid efficacy and the half-life of the colloid would then be reduced.
Can the tracer studies still be correct? Yes, they can. Lamke and Liljedahl2 found that isotonic saline expanded the plasma volume by 170 ml when measured 15 min after the infusion of 1000 ml of isotonic saline over 90 min in the postoperative period (Fig. 2). Using volume kinetic data for isotonic saline, the expected volume expansion would arrive at 195 to 200 ml. It is important to understand that 15 min after the completion of a fluid infusion, fluid mechanics are in a nonsteady state when fluid efficacy is changing rapidly.
Do we need colloids?
The downside of colloid fluids is that they give rise to adverse effects (allergy, coagulopathies) not shared by crystalloids. A claimed benefit is high fluid efficacy. As we have seen here, their efficacy is not superior to crystalloids for short (<15 min) and long-lasting (>12 h) infusions, although superiority within this time frame is apparent.
Another claimed benefit of colloids is the production of less interstitial oedema. Even this argument may be questioned, as the plasma volume expansion subsides due to capillary leak, which subsequently increases the interstitial fluid volume in the same way as crystalloids do (albeit more slowly). Oedema will develop if the ratio between capillary leak and fluid-induced urinary excretion exceeds 1.0. Fortunately, this ratio is very close to 1.0 when albumin 5% or HES 130/0.4 is infused in volunteers,14,15 but the diuretic response to colloid fluid in surgical and ICU settings is still unknown.
In the aftermath of the findings of an association between HES and renal impairment,7 there follows a suspicion that hyperoncoticity might develop when the plasma volume expansion has subsided. Most colloid preparations are iso-oncotic but, at least for albumin 5% or HES 130/0.4, the half-life of the macromolecules in the body is normally several times longer than for the accompanying fluid volume. As the HES molecules undergo sequential breakdown, the oncotic pressure is also likely to decrease more slowly than that indicated by their half-life. We must remember that the half-life of starch is derived from weight/volume measurements and not from the number of macromolecules, which is what governs the oncotic pressure.
Assistance with the study: none declared.
Financial support and sponsorship: none declared.
Conflicts of interest: none declared.
Comment from the Editor: this article was checked and accepted by the editors, but was not sent for external peer-review.
1. Moore FD, Dagher FJ, Boyden CM, et al Hemorrhage in normal man: I. Distribution and dispersal of saline infusions following acute blood loss. Ann Surg
2. Lamke LO, Liljedahl SO. Plasma volume changes after infusion of various plasma expanders. Resuscitation
3. Koomans HA, Geers AB, Bier P, et al A study of the distribution of body fluids after rapid saline expansion in normal subjects and in patients with renal insufficiency: preferential intravascular deposition in renal failure. Clin Sci
4. Jacob M, Chappell D, Hofmann-Kiefer K, et al The intravascular volume effect of Ringer[Combining Acute Accent]s lactate is below 20%: a prospective study in humans. Crit Care
5. Strunden MS, Heckel K, Goetz AE, Reuter DA. Perioperative fluid and volume management: physiological basis, tools and strategies. Ann Intensive Care
6. Senagore AJ, Emery T, Luchtefeld M, et al Fluid management for laparoscopic colectomy: a prospective, randomized assessment of goal-directed administration of balanced salt solution of hetastarch coupled with an enhanced recovery program. Dis Col Rectum
7. Perner A, Haase N, Guttormsen AB, et al 6S Trial Group; Scandinavian Critical Care Trials GroupHydroxyethyl starct 130/0.42 versus Ringer[Combining Acute Accent]s acetate in severe sepsis. N Engl J Med
8. Riddez L, Hahn RG, Brismar B, et al Central and regional hemodynamics during acute hypovolemia and volume substitution in volunteers. Crit Care Med
9. Hahn RG. Volume kinetics of infusion fluids (review). Anesthesiology
10. Hahn RG. Volume effect of Ringer[Combining Acute Accent]s solution in the blood during general anaesthesia. Eur J Anaesthesiol
11. Swan H. Experimental acute hemorrhage. Arch Surg
12. Drobin D, Hahn RG. Time course of increased haemodilution in hypotension induced by extradural anaesthesia. Br J Anaesth
13. Li Y, Zhu HB, Zheng X, et al Low doses of esmolol and phenylephrine act as diuretics during intravenous anesthesia. Crit Care
14. Hedin A, Hahn RG. Volume expansion and plasma protein clearance during intravenous infusion of 5% albumin and autologous plasma. Clin Sci
15. Hahn RG, Bergek C, Geback T, Zdolsek J. Interactions between the volume effects of hydroxyethyl starch 130/0.4 and Ringer's acetate. Crit Care, in press