Secondary Logo

Journal Logo

Original Article

Intra- and extravascular volume status in patients undergoing mitral valve replacement: crystalloid vs. colloid priming of cardiopulmonary bypass

Rex, S.*; Scholz, M.; Weyland, A.; Busch, T.; Schorn, B.#; Buhre, W.§

Author Information
European Journal of Anaesthesiology: January 2006 - Volume 23 - Issue 1 - p 1-9
doi: 10.1017/S0265021505001687
  • Free



Cardiopulmonary bypass (CPB) may induce alterations in intravascular volume status and extravascular body water content [1,2], leading to oedema and cardiopulmonary dysfunction [3,4]. With the onset of CPB, the blood is diluted by the prime volume of the extracorporeal circuit resulting in a decrease of colloid osmotic pressure (COP) [5-7]. CPB causes increased capillary permeability by activation of inflammatory mediator systems [8-10]. Furthermore, the hydrostatic pressure within the capillaries is influenced by hypothermia [11] and postischaemic myocardial dysfunction as a consequence of altered myocardial lymphatic function after cardioplegic arrest [12]. Thus, net filtration of fluid into the interstitium is increased during and after extracorporeal circulation, resulting in elevated extravascular lung water content [4,13] and oedema formation in the myocardial interstitium [3,14] which can contribute to ventricular dysfunction. However, it is still questionable if the negative side-effects of CPB can be influenced by the composition of the priming solution [4,6,7,13,15-17]. Both crystalloid and colloid priming solutions are still in use and no standard has gained general acceptance until now [18-20].

Acute changes in intra- and extravascular fluid balance, intravascular volume status and cardiac function have been studied mainly in patients undergoing coronary artery bypass surgery or in mixed patient populations. However, these patients are at a low risk of developing cardiac or pulmonary complications in the perioperative course. No randomized studies have been performed in patients undergoing mitral valve surgery, which have an increased risk to develop fluid accumulation, myocardial oedema and cardiac dysfunction after CPB because of pulmonary hypertension and pre-existing myocardial dysfunction [21,22].

The present study was designed to analyse the effects of crystalloid vs. colloid priming on central haemodynamics, intra- and extravascular volume status and cardiac function inpatients with mitral valve insufficiency prior and up to 24 h after mitral valve replacement.


After approval by the institutional review board committee and written informed consent, 22 patients undergoing, elective mitral valve replacement participated in the study. All patients suffered from mitral valve insufficiency degree III or IV. Patients with an ejection fraction <30% were excluded from the study.

The patients received 1-2 mg flunitrazepam orally in the evening before surgery and 1 h before arrival in the operating room. Preoperative cardiac medication was continued until the day of surgery, except for diuretics and digitalis. Prior to induction of anaesthesia, routine haemodynamic monitoring was established consisting of electrocardiography (leads II and V5) and continuous measurement of central venous and arterial pressure. For this purpose a central venous catheter (CVC) was placed in the right internal jugular vein and a 5-F introducer sheath was inserted into the right femoral artery. General anaesthesia was induced with fentanyl 7 μg kg−1 and etomidate 0.3 mg kg−1. Pancuronium 0.1 mg kg−1 was administered to facilitate tracheal intubation. Anaesthesia was maintained by continuous intravenous (i.v.) infusion of 7 μg kg−1 h−1 fentanyl and midazolam 100 μg kg−1 h−1. After intubation, tidal volume and inspired oxygen concentration (FiO2) were adjusted to maintain normal arterial partial pressure of carbon dioxide (PaCO2) levels and arterial oxygen saturation greater than 95%. During anaesthesia, a 4-F fiberoptic-thermistor catheter was inserted via the introducer sheath in the femoral artery and advanced to the level of the diaphragm. A 7-F pulmonary artery catheter was introduced via the right internal jugular vein. For measurements of indicator dilution curves, the pulmonary artery catheter and the fiberoptic-thermistor catheter were connected to an opto-electronic device (COLD system; Pulsion Medical Systems, Munich, Germany), which enables simultaneous detection of thermo- and dye-dilution curves.

Patients were randomly allocated into two groups. In one group (crystalloid priming group (RL), n = 11), 2000 mL of crystalloid solutions were used for priming of CPB. The second group (colloid priming group (HA), n = 11) also received a priming volume of 2000 mL, but 400 mL of Ringer's lactate were replaced by 400 mL of 20% human albumin resulting in an effective albumin concentration of 4%. Fluid therapy and positive inotropic support in the operating room and the intensive care unit (ICU) was performed by anaesthetists blinded to the type of priming solutions and the results of indicator dilution measurements except for cardiac output. Perioperative treatment was standardized along with the following guidelines:

  1. Basic fluid substitution was performed with 1 mL kg−1 h−1 balanced crystalloid solutions.
  2. Haemodynamic stability was defined as a cardiac index (CI) >2.0 L min−1 m−2 and mean arterial pressure > 60 mmHg.
  3. Dobutamine was administered when haemodynamic stabilization could not be reached by adequate fluid replacement or when there were signs of impaired cardiac contractility.
  4. Diuresis of 0.5-1 mL kg−1 h−1 was achieved in the perioperative period, if needed with furosemide.
  5. Packed red blood cells were given when the patients haemoglobin content was below 80g L−1.
  6. Fresh frozen plasma or platelets were only administered when laboratory examinations revealed coagulation disorders.

Extracorporeal circulation was performed in moderate hypothermia (30°C nasopharyngeal temperature) with a non-pulsatile flow of 2.0-2.4 L min−1 m2. CPB consisted of a centrifugal pump (Sarns Delphin; 3M Health Care, Ann Arbor, MI, USA) and a membrane oxygenator (Bard HF 5000; W. Harvey, Tewksbury, MA, USA). Cardiac arrest and myocardial protection were induced by 2000 mL of Bretschneider's cardioplegic solution (Custodiol; Köhler Chemie, Alsbach-Hähnlein, Germany) into the aortic root immediately after aortic cross-clamping. Vena cava superior and inferior were cannulated separately, and cardioplegic solution was suctioned via a right ventricular incision. During CPB, no additional colloid was administered. Routine haemodynamic variables heart rate (HR), mean arterial pressure, central venous pressure (CVP), mean pulmonary artery pressure and pulmonary capillary wedge pressure were obtained together with the indicator dilution measurements after induction of anaesthesia and 1, 6 and 24 h after surgery. Arterial and mixed venous blood samples were obtained for determination of haemoglobin content, oxygen saturation (OSM-3 hemoximeter; Radiometer, Brønshøj, Denmark), blood gas analysis (IL 1302 pH/ blood gas analyser; Instrumentation Laboratory Inc., Lexington, MA, USA), COP (IL-186 Weil Onkometer; Instrumentation Laboratory Inc., Lexington, MA, USA; semipermeable membrane with 95% rejection of albumin) and total protein content (124281; Boehringer Mannheim Diagnostica, Mannheim, Germany).

Measurements of cardiac output, central and pulmonary blood volume and right ventricular end-diastolic volume were performed by triple bolus injections of 10 mL of ice-cooled solution of indocyanine green (2.25 mg mL−1) into the right atrium as previously described in more detail [4,23,24]. The injections were spread randomly over the respiratory cycle. Thermodilution curves in the pulmonary artery and aortic thermo-dye dilution were recorded simultaneously after each bolus injection. Flow-derived and volumetric parameters were normalized to body surface area. CI was assessed from the pulmonary artery thermodilution curve according to the Stewart- Hamilton principle. Central blood volume index (CBVI) was calculated as the product of CI and central mean transit time of the intravascular dye (mttcent-dye), that is the mean transit time of the dye between the pulmonary artery and the aorta [23].

Pulmonary blood volume index (PBVI) was calculated from the product of CI and the exponential decay time (tdt-dye) of the dye curve obtained in the aorta. The decay time can be assessed by the reciprocal value of the time constant characterizing the downslope of the indicator dilution curve [4].

Likewise, right ventricular end-diastolic volume index (RVEDVI) was calculated as the product of CI and the time constant of the downslope of the pulmonary thermodilution curve (tdt-thAP) [24].

Measurement of total blood volume

Measurements of total blood volume (TBV) were achieved by spectrophotometric detection of indocyanine green plasma levels [23,25]. Mixed venous blood samples were obtained before (blank sample) as well as 3, 4, 5, 6, 9, 12, 15, 18, 21, 27 and 30 min after bolus injections of 22.5 mg indocyanine green into the right atrium. The blood samples were analysed for haemoglobin content and plasma was separated by centrifugation with 4000 U min−1 for 10 min. Absorption spectra of the indocyanine green containing plasma samples were recorded between 600 and 900 nm with an spectral analyser (CCD-camera; Theta Systems, Munich, Germany). Indocyanine green plasma concentrations were then determined by multilinear fitting of the entire spectrum to a calibration spectrum. Individual calibration was performed for each measurement of TBV by titrating a known amount of indocyanine green in the blank blood sample. The concentration time course (c(t)) of indocyanine green was described with a non-linear least square fitting procedure using a biexponential model function:

where a and b are the weighing factors and k1 and k2 are the time constants.

The virtual indocyanine green concentration at time of injection (C0) can only be determined by back-extrapolation as complete mixing of indocyanine green in the volume of distribution needs several minutes. C0 was therefore assumed to be:

TBV was then calculated according to the principle that the mass of the injected indicator is conserved in its volume of distribution:

where m0 is the amount of indocyanine green and BW is the body weight.

Fluid balance

Input and output of fluids were recorded and cumulative net fluid balances (NFBs) were calculated at the end of surgery, 1, 6 and 24 h after surgery as previously described [25]. Crystalloid infusions, colloid infusion and erythrocyte volume were recorded separately. The priming volume of CPB was part of the fluid input. Packed red blood cells consisted of an erythrocyte volume of 60%, a colloid volume of 39% and a water content of 1%. The residual volume of the pump circuit at end of surgery and the postoperative blood losses were separated into lost erythrocyte volume (calculated with the corresponding haematocrit value at each time point) and plasma volume which contributed to the colloid output. Before and after CPB, blood losses were collected with a cell-saver and added to the fluid output as described above. The collected blood was washed with normal saline solution and transfused postoperatively after determination of the haematocrit value in the processed blood. For the crystalloid output, water losses by gastric tubes and diuresis were measured directly whereas insensibile loss was estimated (25 mL h−1 during surgery and spontaneous breathing 12.5 mL h−1 during ventilation). Changes in extravascular fluid content (EVFC) were calculated as the cumulative NFB corrected for changes in TBV (ΔTBV) at end of surgery and 1, 6 and 24 h postoperatively:


On the basis of a previous study, in which we analysed the effects of different priming solutions on CBV, a sample size of 20 patients (10 per group) was estimated [26]. Power analysis revealed a sample size of eight patients in each group for a 25% effect on volumetric indices, when a level of significance of 0.05% and a power of 80% were to be achieved. All data in tables and figures are presented as mean ± standard deviation. All data were statistically analysed using a commercially available software package (Statistica© for Windows Version 6.0; Statsoft, Tulsa, OK, USA). Two-group comparison concerning biometric and patient characteristics data was done with t-test or U-test, if appropriate. Differences within group and between groups were tested with analysis of variance for repeated measurements (ANOVA). A level of P < 0.05 was considered significant. In case of significant differences, post-hoc testing was performed using the Tukey honest significant difference test and again a level of P < 0.05 was considered significant.

Data concerning the perioperative time course of extravascular lung water and pulmonary gas exchange in these patients has been published [27].


Both groups were comparable with respect to biometric data, duration of CPB and aortic cross-clamping (Tables 1 and 2). The degree of mitral valve insufficiency and left ventricular function were also comparable between the two groups (Table 1). The type of priming volume did not influence the time until extubation or duration of ICU stay.

Table 1
Table 1:
Biometric data and data comparing duration of CPB, duration of aortic cross clamping and time to extubation.
Table 2
Table 2:
Colloid osmotic pressure (COP), total protein (TP) and haemoglobin (Hb) content during the perioperative time course.

Haemodynamic data

Haemodynamics at baseline did not differ between groups (Table 3). HR was significantly increased in both groups during the entire postoperative period, whereas mean arterial pressure, mean pulmonary artery pressure, pulmonary artery occlusion pressure and pulmonary vascular resistance index remained unchanged. CVP was increased 24 h after surgery in the RL group, but not in the HA group. Compared to baseline, CI was significantly increased in the HA group during the entire postoperative phase, but only at 6 h after surgery in the RL group. A significant decrease of SVI was observed in the early postoperative period only in patients receiving crystalloid priming. However, haemodynamic changes and the need for dobutamine, administered during the study period was not different between groups. Both groups were also comparable concerning volumetric haemodynamic indices (Table 4). Pulmonary blood volume remained at similar values, whereas CBVI was decreased throughout the whole postoperative time course. In contrast, RVEDVI increased during the entire postoperative period.

Table 3
Table 3:
Haemodynamic variables during the perioperative time course.
Table 4
Table 4:
Volumetric indices during the perioperative time course.

Perioperative fluid balance and TBV

In both groups a markedly positive total fluid balance was found at the EOS (Table 5). The positive fluid balances decreased only late in the postoperative period, that is > 6 h after surgery, but remained still positive at the end of the study (10 ± 23 mLkg−1 and 44 ± 20 mL kg−1 in the HA and RL groups, respectively (P ≤ 0.01)). During the study period, total fluid balances were significantly, up to four times, higher in the RL than in the HA group. Patients with crystalloid priming received significantly more crystalloid infusions during surgery than patients with colloid priming. The difference in the cumulative balances between groups continued to be significant until 24 h after surgery. Colloid balances were comparable in both groups. After surgery, the amount and the type (i.e. colloid or crystalloid) of fluid input was comparable in both groups. Postoperatively, crystalloids were replaced by colloids in both groups. There were no significant differences between the groups concerning the dosages of furosemide. Blood loss and the amount of transfused blood was not significantly different between groups in the intra- as well as in the postoperative period.

Table 5
Table 5:
Cumulative net fluid balances in the perioperative time course.

Laboratory values after induction of anaesthesia were similar in both groups (Table 2). A decrease in COP and total protein content was found 1 h after surgery. In patients with colloid priming, COP and total protein content returned to baseline 6 h after surgery. In patients receiving crystalloid priming, COP and total protein values reached control levels after 24 h. There were no statistical differences between the groups at any time point. The perioperative time course of haemoglobin concentration was not influenced by the choice of the priming solution. In both groups, haemoglobin was significantly decreased 6 and 24 h after end of surgery.

TBV was decreased in the RL group at all postoperative time points (Fig. 1) whereas in the HA group, an increase of TBV could be detected which was statistically significant 6 h after surgery. In both groups, an increase of EVFC could be observed which decreased only slowly during the postoperative time course (Fig. 2). In patients receiving colloid priming, significantly less fluid extravasation could be found.

Figure 1.
Figure 1.:
Perioperative time course of TBV after induction of anaesthesia (control), 1, 6, and 24 h after surgery. HA: colloid priming group; RL: crystalloid priming group. Mean ± SD;*;P ≤ 0.05 and**P ≤ 0.01 vs. control;°P ≤ 0.05: RL vs. HA.
Figure 2.
Figure 2.:
Cumulative absolute changes in EVFC from induction of anaesthesia (control) until 24 h postoperatively. HA: colloid priming group; RL: crystalloid priming group. Mean ± SD;°°P ≤ 0.01: RL vs. HA.


The results of the present study suggest that in patients with pure crystalloid priming of CPB, significantly more fluid was transferred to the extravascular space in the intra- and postoperative period, accompanied by a decrease of TBV and an increase in EVFC. To maintain adequate haemodynamics, patients with crystalloid priming required a significantly higher amount of volume substitution intraoperatively, resulting in markedly increased positive fluid balances.

Fluid accumulation in the extravascular space is a common phenomenon associated with CPB [1]. A filtered volume of nearly 3L h−1 has been estimated for a 70-kg person during the period of extracorporeal circulation [2]. Fluid extravasation leads to increased water content in tissues and organs, possibly resulting in cardiac, circulatory and pulmonary dysfunction and peripheral oedema. Transvascular fluid shifts can, in general, be described with Starling's hypothesis of transcapillary forces. However, the mechanisms contributing to the increase of fluid extravasation associated with the use of CPB are complex and only partially understood: (1) a decrease of plasma COP at the onset of CPB due to haemodilution [1,28]; (2) a systemic inflammatory response secondary to the exposure of blood to the foreign surfaces of the CPB circuit, leading to capillary leakage [8,9]; (3) hypothermia per se due to a change in the fluid filtration coefficient and the capillary hydrostatic pressure [11]; (4) postischaemic impairment of cardiac function resulting in an elevation of capillary hydrostatic pressure; and (5) altered myocardial lymphatic transport characteristics [12].

The present study differed from previous studies in several important aspects: first, we directly measured intravascular and intracardiac volumes (i.e. CBV, PBV and TBV) in order to quantify changes in volume status. In most other studies, indirect parameters like CVP, pulmonary artery occlusion pressure and cardiac output were used as substitutes for intravascular volume status. Second, we performed detailed fluid balances that enabled us, in combination with the assessment of TBV, to calculate changes in EVFC. Finally, the majority of studies were performed in patients undergoing coronary artery bypass grafting (CABG), whereas we studied patients with pre-existing mitral valve insufficiency, a subgroup of patients with increased perioperative risk of fluid accumulation.

As expected, we found a marked increase in EVFC after CPB. The rise of EVFC was significantly influenced by the choice of the priming solution: in patients with colloid priming, the increase in EVFC 1 h after surgery was only half of that seen in patients with pure crystalloid priming (28 ± 19 mL kg−1 vs. 63 ± 19 mL kg−1, P < 0.01). Moreover, EVFC returned to baseline 24 h after surgery in the HA group whereas in the RL group, EVFC remained significantly elevated. This is in accordance with studies, in which the use of a colloid priming also led to less fluid extravasation [5-7,29]. During CPB, extravasation of fluid leads to depletion of intravascular volume so that additional crystalloids are required to maintain the volume of the venous reservoir, pump flow and intravascular volume load. This explains why patients in the RL group received significantly more crystalloids intraoperatively (in an amount exceeding the priming volume) and had markedly higher total fluid balances than patients in the HA group [5,16,17].

Recently, the concept of an inflammatory mechanism which results in an increase in capillary permeability, not only to aqueous solutions, but also to macromolecules like albumin has been questioned [30-32]. However, although protein and albumin extravasation were not determined in our study, a significant protein shift from the intra- to the extravascular space cannot be ruled out. In both groups, the increase in EVFC exceeded not only the net balance of crystalloid solutions 1 h after surgery, but was approximately equivalent to the sum of both crystalloid and colloid input. Our results thus demonstrate that not only the main portion of the infused crystalloids was transferred to the extravascular space but that there was also an extravasation of albumin molecules. Similar observations have recently been reported in septic, critically ill patients [33]. In these patients, 5% albumin and normal saline increased interstitial fluid volume by an amount approximately equivalent to the infused volume. As the inflammatory response reaches a maximum 2-4 h after CPB [34], it is comprehensible that the postoperative infusion of colloid solutions in both groups did not lead to a further extravasation of macromolecules but to an elevation of plasmatic COP and thus, along with the use of diuretics, resulted in a trend towards lower EVFC. Similar observations were made by Ernest and colleagues [35], who found no expansion of interstitial fluid after postoperative infusion of 5% albumin in cardiac surgical patients. However, as intravascular fluid loss was higher in the RL group than in patients receiving colloid priming, it seems likely that a second mechanism beyond the inflammatory reaction contributes to fluid extravasation. The decrease in plasma COP did not differ significantly in our groups at the chosen time points. This is in contrast to other studies, in which COP was significantly lower when Ringer's lactate was used instead of albumin, hydroxyethyl starch or gelatin. In these investigations, the difference in COP was only present during surgery; similar to our findings, COP was comparable between groups, 1 h and later after surgery and the use of colloid priming was associated with less fluid accumulation [5,7,15]. Although we did not measure COP during CPB, it is most likely that the decrease of COP during CPB was less pronounced in patients receiving a nearly isooncotic priming. This could have led to significantly less fluid extravasation during surgery in the HA group.

The perioperative time course of TBV in the HA group is comparable to that described by Hoeft and colleagues in CABG patients [23]. Less fluid extravasation in the HA group was associated with an increase in TBV. In contrast, patients receiving pure crystalloid priming showed a significant decrease of TBV despite markedly positive fluid balances which were obviously not able to compensate for the increased fluid extravasation. One could assume that an increase in TBV is associated with an increase in cardiac preload and performance. However, not TBV but the intravascular volume in the intrathoracic compartment (i.e. CBV) was shown to be the most reliable parameter of left ventricular preload as it represents the reservoir for the left ventricle [36]. In both groups, CBVI was significantly decreased after surgery. After substitution with colloid solutions in the immediate postoperative period, CBVI returned to baseline, accompanied by an increase in SVI in patients receiving pure crystalloid priming. A general increase in EVFC theoretically leads also to an increase in the water content of the myocardial interstitium. Thus, patients in the RL group should have suffered from a more pronounced myocardial oedema than patients in the HA group. In patients undergoing surgery for CABG, either no or only insignificant depression of cardiac performance was observed when a pure crystalloid priming was used [7,16,17,25,29]. In the present study, SVI was significantly decreased in the immediate postoperative period in patients receiving pure crystalloid priming (30 ± 6 vs. 40 ± 7 mL m−2, P < 0.01) when compared to baseline. However, cardiac function parameters and the amount of positive inotropic support were comparable in both groups during the whole perioperative period, suggesting that the choice of priming volume did not influence clinical outcome parameters. However, one of the limitations of this study is that the study size is too small for the analysis of outcome parameters.

In summary, our results demonstrate that both colloid and crystalloid CPB priming solutions were transferred to the extravascular space. However, patients with colloid priming of CPB showed significantly less fluid extravasation, had decreased positive fluid balances and required less intraoperative fluid replacement.


Wolfgang Buhre has received honoraria for lectures from Pulsion Medical Systems, Germany, the Manufacturer of the COLD-System. Wolfgang Buhre is also a member of the Medical Advisory Board of Pulsion Medical Systems.


1. Olthof CG, Jansen PG, de Vries JP et al. Interstitial fluid volume during cardiac surgery measured by means of a non-invasive conductivity technique. Acta Anaesthesiol Scand 1995; 39: 508-512.
2. Koller ME, Bert J, Segadal L, Reed RK. Estimation of total body fluid shifts between plasma and interstitium in man during extracorporeal circulation. Acta Anaesthesiol Scand 1992; 36: 255-259.
3. Geissler HJ, Allen SJ. Myocardial fluid balance: pathophysiology and clinical implications. Thorac Cardiovasc Surg 1998; 46 (Suppl 2): 242-245.
4. Hachenberg T, Tenling A, Rothen HU, Nystrom SO, Tyden H, Hedenstierna G. Thoracic intravascular and extravascular fluid volumes in cardiac surgical patients. Anesthesiology 1993; 79: 976-984.
5. Riegger LQ, Voepel-Lewis T, Kulik TJ et al. Albumin versus crystalloid prime solution for cardiopulmonary bypass in young children. Crit Care Med 2002; 30: 2649-2654.
6. Eising GP, Niemeyer M, Gunther T et al. Does a hyperoncotic cardiopulmonary bypass prime affect extravascular lung water and cardiopulmonary function in patients undergoing coronary artery bypass surgery? Eur J Cardiothorac Surg 2001; 20: 282-289.
7. Sade RM, Stroud MR, Crawford Jr FA, Kratz JM, Dearing JP, Bartles DM. A prospective randomized study of hydroxyethyl starch, albumin, and lactated Ringer's solution as priming fluid for cardiopulmonary bypass. J Thorac Cardiovasc Surg 1985; 89: 713-722.
8. Levy JH, Tanaka KA. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 2003; 75: S715-S720.
9. Wan S, LeClerc JL, Vincent JL. Inflammatory response to cardiopulmonary bypass: mechanisms involved and possible therapeutic strategies. Chest 1997; 112: 676-692.
10. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW. Complement activation during cardiopulmonary bypass: evidence for generation of C3a and C5a anaphylatoxins. New Engl J Med 1981; 304: 497-503.
11. Heltne JK, Bert J, Lund T et al. Temperature-related fluid extravasation during cardiopulmonary bypass: an analysis of filtration coefficients and transcapillary pressures. Acta Anaesthesiol Scand 2002; 46: 51-56.
12. Mehlhorn U, Davis KL, Burke EJ, Adams D, Laine GA, Allen SJ. Impact of cardiopulmonary bypass and cardioplegic arrest on myocardial lymphatic function. Am J Physiol 1995; 268: H178-H183.
13. Hoeft A, Korb H, Mehlhorn U, Stephan H, Sonntag H. Priming of cardiopulmonary bypass with human albumin or Ringer lactate: effect on colloid osmotic pressure and extravascular lung water. Br J Anaesth 1991; 66: 73-80.
14. Mehlhorn U, Allen SJ, Adams DL et al. Normothermic continuous antegrade blood cardioplegia does not prevent myocardial edema and cardiac dysfunction. Circulation 1995; 92: 1940-1946.
15. Jansen PG, te Velthuis H, Wildevuur WR et al. Cardiopulmonary bypass with modified fluid gelatin and heparincoated circuits. Br J Anaesth 1996; 76: 13-19.
16. Scott DA, Hore PJ, Cannata J, Masson K, Treagus B, Mullaly J. A comparison of albumin, polygeline and crystalloid priming solutions for cardiopulmonary bypass in patients having coronary artery bypass graft surgery. Perfusion 1995; 10: 415-424.
17. Marelli D, Paul A, Samson R, Edgell D, Angood P, Chiu RC. Does the addition of albumin to the prime solution in cardiopulmonary bypass affect clinical outcome? A prospective randomized study. J Thorac Cardiovasc Surg 1989; 98: 751-756.
18. Boldt J. Volume therapy in cardiac surgery: does the kind of fluid matter? J Cardiothorac Vasc Anesth 1999; 13: 752-763.
19. London MJ. Pro: colloids should be added to the pump prime. J Cardiothorac Anesth 1990; 4: 401-405.
20. D'Ambra MN, Philbin DM. Con: colloids should not be added to the pump prime. J Cardiothorac Anesth 1990; 4: 406-408.
21. Mehlhorn U, Allen SJ, Davis KL, Geissler HJ, Warters RD, Rainer de Vivie E. Increasing the colloid osmotic pressure of cardiopulmonary bypass prime and normothermic blood cardioplegia minimizes myocardial oedema and prevents cardiac dysfunction. Cardiovasc Surg 1998; 6: 274-281.
22. Davis KL, Mehlhorn U, Laine GA, Allen SJ. Myocardial edema, left ventricular function, and pulmonary hypertension. J Appl Physiol 1995; 78: 132-137.
23. Hoeft A, Schorn B, Weyland A et al. Bedside assessment of intravascular volume status in patients undergoing coronary bypass surgery. Anesthesiology 1994; 81: 76-86.
24. Lichtwarck-Aschoff M, Leucht S, Kisch HW, Zimmermann G, Blumel G, Pfeiffer UJ. Monitoring of right ventricular function using a conventional slow response thermistor catheter. Intens Care Med 1994; 20: 348-353.
25. von Spiegel T, Giannaris S, Wietasch GJ et al. Effects of dexamethasone on intravascular and extravascular fluid balance in patients undergoing coronary bypass surgery with cardiopulmonary bypass. Anesthesiology 2002; 96: 827-834.
26. Buhre W, Weyland A, Schorn B et al. Changes in central venous pressure and pulmonary capillary wedge pressure do not indicate changes in right and left heart volume in patients undergoing coronary artery bypass surgery. Eur J Anaesthesiol 1999; 16: 11-17.
27. Buhre W, Hoeft A, Schorn B, Weyland A, Scholz M, Sonntag H. Acute affect of mitral valve replacement on extravascular lung water in patients receiving colloid or crystalloid priming of cardiopulmonary bypass. Br J Anaesth 1997; 79: 311-316.
28. English TA, Kirklin JW, Digerness S. Changes in colloid osmotic pressure during and shortly after open intracardiac operation. J Thorac Cardiovasc Surg 1971; 61: 338-341.
29. Jarvela K, Koskinen M, Kaukinen S, Koobi T. Effects of hypertonic saline (7.5%) on extracellular fluid volumes compared with normal saline (0.9%) and 6% hydroxyethyl starch after aortocoronary bypass graft surgery. J Cardiothorac Vasc Anesth 2001; 15: 210-215.
30. Heltne JK, Koller ME, Lund T et al. Studies on fluid extravasation related to induced hypothermia during cardiopulmonary bypass in piglets. Acta Anaesthesiol Scand 2001; 45: 720-728.
31. Tassani P, Schad H, Winkler C et al. Capillary leak syndrome after cardiopulmonary bypass in elective, uncomplicated coronary artery bypass grafting operations: does it exist? J Thorac Cardiovasc Surg 2002; 123: 735-741.
32. Falk JL, Rackow EC, Weil MH. Colloid and crystalloid fluid resuscitation. Acute Care 1983; 10: 59-94.
33. Ernest D, Belzberg AS, Dodek PM. Distribution of normal saline and 5% albumin infusions in septic patients. Crit Care Med 1999; 27: 46-50.
34. Levy JH, Kelly AB. Inflammation and cardiopulmonary bypass. Can J Anaesth 1993; 40: 1009-1015.
35. Ernest D, Belzberg AS, Dodek PM. Distribution of normal saline and 5% albumin infusions in cardiac surgical patients. Crit Care Med 2001; 29: 2299-2302.
36. Arndt JO. The low pressure system: the integrated function of veins. Eur J Anaesthesiol 1986; 3: 343-370.


© 2006 European Society of Anaesthesiology