BJA Advance Access originally published online on December 24, 2007
British Journal of Anaesthesia 2008 100(3):307-314; doi:10.1093/bja/aem363
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Effects of colloid and crystalloid solutions on endogenous activation of fibrinolysis and resistance of polymerized fibrin to recombinant tissue plasminogen activator added ex vivo
1 Department of Anaesthesiology and Intensive Care Medicine
2 Department of Pediatrics, Innsbruck Medical University, Anichstrasse 35, A-6020 Innsbruck, Austria
3 Assign Data Management and Biostatistics GmbH, A-6020 Innsbruck, Austria
* Corresponding author. E-mail: markus.mittermayr{at}i-med.ac.at
Accepted for publication September 11, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background: The study was conducted to explore the effects of colloid and crystalloid solutions on activation of fibrinolysis during orthopaedic surgery and to determine whether fluids facilitate clot dissolution at a particular fibrinolytic activity.
Methods: Tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) were measured in plasma samples of 66 orthopaedic patients randomly receiving gelatin solution, hydroxyethyl starch (HES) (130/0.4), or exclusively Ringer’s lactate solution. Plasma obtained before induction of anaesthesia (undiluted) and at the end of surgery (diluted) was exposed to recombinant tissue plasminogen activator (r-tPA) in vitro and analysed by modified thrombelastography (ROTEM®).
Results: There were similar changes in t-PA and PAI-1 concentrations in the gelatin, HES, and Ringer’s lactate groups. When compared with the effect of r-tPA on undiluted plasma samples, the presence of colloids prompted faster clot dissolution than did Ringer’s lactate solution. Lysis index at 30 min decreased significantly [median (min/max); P vs Ringer’s lactate solution] to 43 (1/82)% (P=0.007), 14 (3/70)% (P<0.001), and 91 (34/97)%, lysis onset time decreased to 1269 (1054/1743) s (P=0.007), 972 (490/1565) s (P<0.001), and 1970 (1260/2165) s, and lysis time to 2469 (1586/3303) s (P=0.019), 2002 (1569/3600) s (P=0.006), and 3012 (2017/3600) s in the gelatin, HES, and Ringer’s lactate groups, respectively.
Conclusions: The type of i.v. fluid used does not influence endogenously occurring fibrinolytic activity in patients undergoing major orthopaedic surgery. However, during hyperfibrinolysis, the presence of HES or gelatin solution facilitates clot disintegration to a greater extent than Ringer’s lactate solution, because the weaker clots formed with colloids dissolve faster.
Keywords: blood, coagulation; blood, haemodilution; measurement techniques, thrombelastograph
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In a previous study focusing on the influence of i.v. fluids on procoagulant factor activities and their functional consequences on clot formation, we showed that during major orthopaedic surgery the type of i.v. solution administered influences clot formation mainly by interfering with fibrinogen/fibrin polymerization, with hydroxyethyl starch (HES) 130/0.4 showing the greatest impairment of clot formation, followed by gelatin solution, and finally an exclusively crystalloid regimen showing the least effect.1 The quality of fibrinogen/fibrin polymerization depends on sufficient thrombin generation, concentrations, and functionality of fibrinogen, the activity of coagulation factor XIII (FXIII), and the activity of the fibrinolytic system. In that study, we found no evidence for fluid-specific differences in thrombin generation as evaluated by serial measurements of concentrations of TAT, F1+2, and DD, nor did we find differences in declining plasma concentrations of fibrinogen measured using the Clauss method. Furthermore, we observed that the activity of FXIII was comparable between the groups. In addition, we found no evidence of increased fibrinolytic activity when analysing the lysis index at 30 min; this index measures the percentage of remaining clot firmness at 30 min in relation to the maximum measured clot firmness. However, because these measurements were obtained intermittently and ROTEM tracings were discontinued after 40 min, these data were not appropriate for addressing the frequently asked question ‘whether i.v. fluids interfere with endogenous release of fibrinolytic activity or facilitate dissolution of formed fibrin clots as suggested by data in patients undergoing major abdominal surgery’.2
In human plasma diluted in vitro and spiked ex vivo with t-PA in the presence of albumin and HES 130/0.4, Nielsen3 showed that clots disintegrated faster than when diluted with NaCl 0.9%. Interestingly, rabbits that were haemodiluted in vivo by infusion with various HES preparations showed a markedly enhanced fibrinolytic response in comparison with in vitro-diluted human plasma. To date, it is unclear whether these observed differences relate to the difference in study design (in vitro/in vivo dilution) or are species-related. Moreover, the effects of gelatin solution, which is frequently used to maintain normovolaemia, are unknown.
To test the hypothesis that the in vivo presence of HES or gelatin solution as used in clinical practice in combination with Ringer’s lactate solution facilitates clot disintegration when compared with an exclusively Ringer’s lactate regimen, we analysed plasma samples collected for a previous study. Specifically from orthopaedic patients undergoing surgery of the spine and randomly receiving gelatin, HES 130/0.4, or exclusively Ringer’s lactate solution. The fibrinolytic response after in vitro addition of recombinant tissue plasminogen activator (r-tPA) was assessed using ROTEM® software, thereby providing more detailed data on the kinetics of clot disintegration. Moreover, molecular markers of fibrinolysis were also measured in patients’ plasma to determine whether the endogenously occurring release of activators of fibrinolysis during orthopaedic surgery of the spine is influenced by the choice of fluids administered.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The protocol for the present study was approved by the local University Ethics Committee. Patients were part of a previous investigation of the haemostatic effects of crystalloid and colloid solutions during major orthopaedic surgery. A detailed description of patient selection, standardized general anaesthesia, and monitoring is given in our previously published article.1 In brief, 66 consecutive patients (ASA I–II) undergoing surgery of the spine (more than three segments) were enrolled after obtaining informed written consent for blood sampling.
All patients received propofol, fentanyl, and rocuronium for induction, and anaesthesia was maintained with sevoflurane in an oxygen/air mixture supplemented with an infusion of remifentanil. All patients were monitored invasively, actively warmed, and a blood salvage device was used. As a trigger for transfusion of red cells, a haemoglobin value <8 mg/dl was used. Furthermore, fibrinogen concentrate (Haemocomplettan®, CSL Behring GmbH, Marburg, Germany) was administered in cases showing reduced fibrinogen/fibrin polymerization in the ROTEM® assay [FIBTEM maximum clot firmness (MCF) <7 mm]; FFP would be administered to correct prolonged initiation of coagulation unresponsive to prothrombin complex concentrate and platelets would be transfused if total clot strength remained critically reduced after correcting fibrinogen polymerization.
Bearing in mind the difference in volume effect between the colloid (gelatin solution 70%, HES 100%) and the crystalloid solutions (30%) tested and using a computer-generated randomization list, patients received 8–11 ml kg–1 h–1 modified gelatin solution (4% Gelofusin®, B. Braun, Maria Enzersdorf, Austria) or 6–8 ml kg–1 h–1 medium molecular weight medium-substituted HES (6% Voluven® 130/0.4, Fresenius, Pharma Austria GmbH, Graz, Austria) in addition to a basic infusion of 5 ml kg–1 h–1 Ringer’s lactate solution (Ringer Laktat®, Fresenius Pharma Austria GmbH, Graz, Austria) or exclusively 13–15 ml kg–1 h–1 Ringer’s lactate solution throughout the intraoperative study period. If hypovolaemia was suspected, additional group-specific fluid was administered.
Concentrations of tissue-type plasminogen activator (t-PA) (Zymutest tPA Ag, Hyphen BioMed, Neuville-sur-Oise, France) and plasminogen activator inhibitor-1 (PAI-1) (ELISA PAI-1-Antigen, Haemochrom Diagnostica, Essen, Germany) were measured using the appropriate ELISA technique.
Plasma was separated from arterial blood samples drawn at baseline [before induction of anaesthesia (A)], after 60 min and immediately before surgical incision (B), and every 90 min thereafter (C, D).
Modified thrombelastography (ROTEM®, Pentapharm GmbH, Munich, Germany), using the newly available software (1.0.04-2005-12-15) was performed in plasma samples from the first 10 patients per group, obtained before induction of anaesthesia (time point A, undiluted) and after 240 min (time point D, diluted). ROTEM® analysis was performed in aliquots of 300 µl undiluted and diluted plasma samples, with (+r-tPA) or without (no r-tPA) addition of 0.1 µg r-tPA (Actilyse®, Boehringer Ingelheim Pharma GmbH & Co. KG, Germany) and using one lot for all experiments. To activate coagulation and analysis of the formed fibrinogen/fibrin clot, the extrinsically activated EXTEM test was performed (20 µl of CaCl2 0.2 M, 20 µl tissue factor, 300 µl of plasma). All reagents were purchased from Pentapharm Co., and measurements were performed at 37°C. The ROTEM® devices used were checked for correct function using quality control serum (ROTROL®, Pentapharm Co.).
ROTEM® variables analysed were coagulation time (CT) and MCF. Furthermore, the lysis index at 30 min (LI30), which describes the percentage of clot firmness remaining after 30 min in relation to maximum measured clot firmness, the lysis time (LT), which indicates the time needed for clot firmness to decrease by 90% of MCF, and the lysis onset time (LOT), which is defined as the time needed for clot firmness to decrease by 15% of MCF, were determined. A typical ROTEM® tracing and its interpretation are given in Figure 1.
|
All measurements were run for at least 60 min. For statistical analysis, ‘no fibrinolysis’ was defined as follows: LI30=100%, LT=3600 s, LOT=3600 s.
Statistical analysis
Molecular markers of fibrinolysis
Data are given as median (min, max). Differences in baseline values between the groups were analysed using the Kruskal–Wallis test. To investigate time dependencies, a Friedman ANOVA was applied. The area under the curve was calculated after subtracting baseline values (AUC–BLA–D) and was analysed with the Kruskal–Wallis test and post hoc Wilcoxon’s rank sum tests for comparison of between-group differences in the intraoperative response profile. A P-value of <0.05 was considered statistically significant. The sample size of 18 per group (excluding dropouts) was planned to provide 80% power for detection of a difference in the AUC baseline of –10 vs –50 (SD: ±40) with a two-sided significance level of 5%.
ROTEM assays for resistance to r-tPA
Data are given as medians (min, max). To determine effects of haemodilution within a group, the Wilcoxon test was applied; to determine the effects of haemodilution on CT and MCF without lysis between the groups, the differences between the diluted and undiluted controls (no r-tPA) were analysed using the Kruskal–Wallis test. Differences in baseline values were analysed with the Kruskal–Wallis test. To detect only the effect of lysis in diluted plasma, differences between the groups were analysed by calculating and comparing differences between the diluted (+r-tPA–no r-tPA) and undiluted lysis effect (+r-tPA–no r-tPA) using the Kruskal–Wallis test. In the case of significance, the Mann–Whitney U-test was applied. A P-value of <0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We investigated plasma samples from 66 patients. Five patients were excluded from the study because of unexpected pathological baseline values of platelets and fibrinogen. There were no differences between the groups regarding patient demographics (Table 1) or baseline measurements of molecular markers of fibrinolysis or fibrinogen/fibrin polymerization as measured using the ROTEM assay. Data on intraoperative volume supply are shown in Table 2.
|
|
Molecular markers of fibrinolysis
When compared with baseline, concentrations of t-PA decreased significantly during surgery from time A to time D in the gelatin, HES, and Ringer’s lactate groups (ANOVA; P<0.001, P<0.001, P<0.001, respectively). However, no differences in the intraoperative response profile of molecular markers of fibrinolysis between the groups (comparison of AUC–BLA–D) were observed (Fig. 2A). There was an inconsistent decrease in PAI-1 values from time A to time D in the gelatin, HES, and Ringer’s lactate solution groups (P<0.001, P<0.001, P<0.001, respectively). No between-group differences were observed (Fig. 2B). Moreover, the PAI-1:t-PA ratio as a marker of the state of fibrinolysis showed no differences among the groups (Fig. 2C).
|
In vitro activation of fibrinolysis
The effects of fibrinolytic stress on thrombelastographic variables of clot growth in undiluted and diluted plasma samples are shown in Figure 3.
|
When compared with undiluted samples, haemodilution alone caused a significant increase in CT in the HES group (P=0.005). The increase was not significant in the gelatin group. With Ringer’s lactate, no change was observed. When samples were additionally exposed to r-tPA, the change in CT measurements was comparable between the groups.
MCF decreased significantly more from the undiluted to the diluted control (no-rtPA) when patients received gelatin or HES (P<0.002, P<0.001, respectively) than when they were given Ringer’s lactate solution. When additionally exposed to r-tPA, the further decrease in MCF showed no differences between the groups (Fig. 3).
In undiluted samples, no lysis was detectable by ROTEM® assay. Addition of r-tPA to undiluted samples decreased the lysis index at 30 min (LI30), LOT, and LT.
In diluted samples containing HES, or gelatin, addition of r-TPA decreased the lysis index at 30 min (LI30) (P<0.001, P=0.008 respectively), LOT (P<0.001, P=0.007, respectively), and LT (P=0.006, P=0.02, respectively) significantly more than was observed for samples containing Ringer’s lactate and resulted in faster clot dissolution (Fig. 4).
|
Representative ROTEM® tracings of undiluted plasma or plasma diluted with gelatin, HES, or Ringer’s lactate solution and additionally exposed to r-tPA are depicted in Figure 5.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The main finding of our study is that administration of HES or gelatin, in comparison with an exclusively crystalloid regimen, does not influence the endogenous release of activators of fibrinolysis or the activity of PAI-1 in haemodynamically stable orthopaedic patients undergoing surgery of the spine. However, when such diluted plasma is exposed to considerable fibrinolytic stress, as tested by in vitro addition of r-TPA, the weaker fibrin clots observed in the presence of HES or gelatin dissolve faster as measured by reduced lysis index at 30 min, decreased LOT and LT. Thus, the resulting clot disintegration is faster than for Ringer’s lactate solution. Colloids have been shown to impair coagulation more than crystalloid solutions in many in vitro and in vivo investigations.4–8 In a previous study, the results of which prompted the present analysis, we demonstrated that impaired fibrinogen/fibrin polymerization is the main problem underlying dilutional coagulopathy during major orthopaedic surgery and that this can be effectively reversed by administering fibrinogen concentrate. In that study, 13 patients in both colloid groups but none in the Ringer’s lactate group developed critically reduced fibrinogen polymerization and thus received fibrinogen concentrate in order to maintain total clot firmness at acceptable levels. This therapeutic intervention most likely explains why estimated and calculated blood losses were comparable among groups, despite the impairment of fibrinogen polymerization frequently observed for both colloids.1
Increased fibrinolytic activity is frequently presumed to aggravate dilutional coagulopathy in various clinical situations such as cardiopulmonary bypass, liver transplantation, traumatized patients, or major surgery with considerable tissue trauma. In such cases, colloids are commonly administered in addition to crystalloid solutions as a means of maintaining normovolaemia. Because surgery of the spine is associated with considerable tissue trauma and patients routinely receive fluids in clinical practice, we decided to perform the present study to more precisely investigate the fibrinolytic process.
We found a continuous decrease in tissue plasminogen activator concentration and a variable decrease in PAI-1, which was independent of the type of fluid administered. Interestingly, despite the more extensive tissue trauma associated with surgery of the spine, observed values were comparable with those reported during knee replacement surgery.9 In contrast, Boldt and colleagues2 demonstrated that administration of HES in patients undergoing major abdominal surgery caused an increase in thrombelastographically measured clot firmness when aprotinin was added in vitro, indicating increased fibrinolysis, which can be antagonized with aprotinin. That study, however, used a special HES preparation, namely Hextend® (HES 550/0.7 in balanced electrolyte solution also containing Ca2+, Mg2+, and glucose), suggesting that results may not translate to other more commonly used HES preparations. Furthermore, that study lacks data on lysis index or LOT and thus its findings are difficult to interpret.
When fibrinolytic activity is enhanced, increased bleeding is commonly observed, which has been reported in several clinical situations (e.g. after cardiopulmonary bypass, treatment with ventricular assist device, in multiple traumatized patients, during orthotopic liver transplantation).10–13 In such situations, the administration of colloids might aggravate haemostatic disturbances and the amount of further blood loss. Kuitunen and colleagues14 assumed that in the presence of HES, fibrin clots are more vulnerable to fibrinolysis, but showed that administration of tranexamic acid after cardiopulmonary bypass failed to reverse the HES-associated impairment of clot strength. However, in that study fibrinolytic activities in patients after cardiopulmonary bypass, as measured by lysis index and maximum lysis, were marginal and did not differ from those of a normal population, suggesting no relevant hyperfibrinolysis was present in such patients. Despite the fact that lysis index at 30, 45, and 60 min after infusion of HES did not differ from baseline, the authors nevertheless concluded that HES has a slightly enhanced effect on fibrinolysis because maximum lysis increased over time but still remained within normal physiological limits (<15%).
Using additional thrombelastographic variables, Nielsen3 found that in human plasma diluted in vitro and spiked ex vivo with t-PA in the presence of albumin and HES 130/0.4, clots disintegrated faster than when diluted with 0.9% of NaCl. Interestingly, rabbits that were haemodiluted in vivo by infusion with various HES preparations showed a markedly enhanced fibrinolytic response when compared with in vitro-diluted human plasma. Using a comparable final concentration of r-tPA, we also found that plasma from patients who had received colloids combined with Ringer’s lactate solution disintegrated faster in the presence of colloids than in the presence of solely Ringer’s lactate. The enhanced dissolution results from the fact that the weaker clots, as seen after colloid administration, also dissolve much faster.
There are several limitations to our study. We cannot exclude the possibility that the additional presence of RL influenced the results in the colloid groups. However, it was our intention to analyse samples from patients undergoing routine fluid management. In clinical practice, crystalloids and colloids are administered in combination in order to compensate fluid deficiencies and to maintain normovolaemia during ongoing blood loss. An exclusively colloid regimen would not translate to clinical routine. As the study was observational, its clinical relevance can be questioned. We wanted to investigate whether commonly used volume regimens alter the release of endogenous fibrinolytic activity, a question still not conclusively answered. In this context, there is above all a lack of data on gelatin solution, which is frequently used in European countries in patients exhibiting blood loss. Our data show that the choice of fluids administered does not influence endogenous fibrinolytic potential and therefore antifibrinolytics are unlikely to influence the impairment of fibrinogen/fibrin polymerization associated with colloids. Furthermore, in patients with suspected or proven hyperfibrinolysis, we feel it is not only necessary to administer antifibrinolytics but also to quickly correct fibrinogen deficiency in order to improve the strength of clots formed, which will then be more resistant to dissolution. Furthermore, the use of crystalloids rather than colloids should be considered in order to facilitate formation of stable clots and thus stop bleeding.
One particular criticism is that we tested fibrinolysis resistance in plasma samples and not in whole blood. Although unlikely, the possibility cannot be excluded that results differ when cellular components (e.g. platelets) are present in their normal condition in whole blood. However, because whole blood assays measure the quality of fibrinogen polymerization after adding a platelet-blocking substance and because fibrinolysis mainly results in dissolution of the formed fibrin clot, investigation of plasma samples and thus of the fibrinogen/fibrin clots seems justified. In addition, analysing plasma enables all samples to be worked up at the same time using one batch of r-tPA, thereby reducing lot-specific differences in activity and limiting r-tPA costs, which are considerable.
In conclusion, this study shows that compared with exclusively Ringer’s lactate administration of HES or gelatin solution does not enhance the endogenous release of t-PA or suppress PAI-1 in patients undergoing surgery of the spine. However, in the case of hyperfibrinolysis, the presence of colloids and thus of already impaired clot formation enhances fibrinolysis as measured by decreased LOT, which in turn indicates that clot disintegration is faster than for clots formed in the presence of Ringer’s lactate. In clinical situations in which hyperfibrinolysis is suspected or even detected by thrombelastography, these findings should be considered. In addition to administering antifibrinolytics, every effort should be made to improve clot strength. Clot strength and clot resistance to fibrinolysis can be better preserved by crystalloid solutions as long as possible to maintain normovolaemia than by administering gelatin or HES solution.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Mittermayr M, Streif W, Haas T, et al. Hemostatic changes after crystalloid or colloid fluid administration during major orthopaedic surgery: the role of fibrinogen administration. Anesth Analg (2007) 105:905–17.
2 Boldt J, Haisch G, Suttner S, et al. Effects of a new modified, balanced hydroxyethyl starch preparation (Hextend) on measures of coagulation. Br J Anaesth (2002) 89:722–8.
3 Nielsen VG. Hemodilution modulates the time of onset and rate of fibrinolysis in human and rabbit plasma. J Heart Lung Transplant (2006) 25:1344–52.[CrossRef][Web of Science][Medline]
4 de Jonge E, Levi M. Effects of different plasma substitutes on blood coagulation: a comparative review. Crit Care Med (2001) 29:1261–7.[CrossRef][Web of Science][Medline]
5 Fries D, Innerhofer P, Klingler A, et al. The effect of the combined administration of colloids and lactated Ringer’s solution on the coagulation system: an in vitro study using thrombelastograph coagulation analysis (ROTEG). Anesth Analg (2002) 94:1280–7. table of contents.
6 Jones SB, Whitten CW, Despotis GJ, et al. The influence of crystalloid and colloid replacement solutions in acute normovolemic hemodilution: a preliminary survey of hemostatic markers. Anesth Analg (2003) 96:363–8. table of contents.
7 Innerhofer P, Fries D, Margreiter J, et al. The effects of perioperatively administered colloids and crystalloids on primary platelet-mediated hemostasis and clot formation. Anesth Analg (2002) 95:858–65. table of contents.
8 Nielsen VG. Colloids decrease clot propagation and strength: role of factor XIII-fibrin polymer and thrombin-fibrinogen interactions. Acta Anaesthesiol Scand (2005) 49:1163–71.[CrossRef][Web of Science][Medline]
9 Fries D, Streif W, Margreiter J, et al. The effects of perioperatively administered crystalloids and colloids on concentrations of molecular markers of activated coagulation and fibrinolysis. Blood Coagul Fibrinolysis (2004) 15:213–9.[CrossRef][Web of Science][Medline]
10 Despotis GJ, Avidan MS, Hogue CW Jr. Mechanisms and attenuation of hemostatic activation during extracorporeal circulation. Ann Thorac Surg (2001) 72:S1821–31.
11 Livingston ER, Fisher CA, Bibidakis EJ, et al. Increased activation of the coagulation and fibrinolytic systems leads to hemorrhagic complications during left ventricular assist implantation. Circulation (1996) 94:II227–34.[Medline]
12 Kapsch DN, Metzler M, Harrington M, et al. Fibrinolytic response to trauma. Surgery (1984) 95:473–8.[Web of Science][Medline]
13 Senzolo M, Burra P, Cholongitas E, et al. New insights into the coagulopathy of liver disease and liver transplantation. World J Gastroenterol (2006) 12:7725–36.[Web of Science][Medline]
14 Kuitunen AH, Suojaranta-Ylinen RT, Kukkonen SI, et al. Tranexamic acid does not correct the haemostatic impairment caused by hydroxyethyl starch (200 kDa/0.5) after cardiac surgery. Blood Coagul Fibrinolysis (2006) 17:639–45.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Bolliger, F. Szlam, J. H. Levy, R. J. Molinaro, and K. A. Tanaka Haemodilution-induced profibrinolytic state is mitigated by fresh-frozen plasma: implications for early haemostatic intervention in massive haemorrhage Br. J. Anaesth., February 3, 2010; (2010): aeq001v1 - aeq001. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Bolliger, F. Szlam, R. J. Molinaro, N. Rahe-Meyer, J. H. Levy, and K. A. Tanaka Finding the optimal concentration range for fibrinogen replacement after severe haemodilution: an in vitro model Br. J. Anaesth., June 1, 2009; 102(6): 793 - 799. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Schramko, R. T. Suojaranta-Ylinen, A. H. Kuitunen, S. I. Kukkonen, and T. T. Niemi Rapidly Degradable Hydroxyethyl Starch Solutions Impair Blood Coagulation After Cardiac Surgery: A Prospective Randomized Trial Anesth. Analg., January 1, 2009; 108(1): 30 - 36. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||