Menzebach and colleagues have introduced a step-by-step approach for the treatment of non-surgical blood loss . We fully agree with the authors that standard instructions for the treatment of diffuse, non-surgical bleeding would be desirable. However, successful management of such patients predominately relies on diagnosis and specific therapy, rather than on a stepwise administration of various haemostatic agents.
The authors suggest the administration of desmopressin (DDAVP) as first-line treatment. DDAVP is clearly beneficial in the case of von Willebrand's disease and mild haemophilia, in patients treated with aspirin preoperatively and in cardiac surgery. Beside these indications, there is no evidence for administering DDAVP for unspecified non-surgical bleeding. A large meta-analysis including 14 trials (1034 patients) showed that DDAVP was not able to minimize perioperative red blood cell concentrate transfusion in patients without any congenital bleeding disorder . Only limited data suggest that this drug may be effective when platelet dysfunction is present. Furthermore, the recommended simultaneous monitoring of International Normalized Ratio is futile. International Normalized Ratio was established to monitor the effects of oral anticoagulation. An increased International Normalized Ratio is neither able to discriminate hypofibrinogenaemia from a deficiency of other plasma clotting factors nor to detect platelet disorders, which can be treated successfully with DDAVP. Although global haemostatic tests as well as platelet count are of limited value, these tests are commonly implemented before any intervention is considered. In addition, the Platelet Function Analyzer (PFA-100®) is well established for the detection of von Willebrand's disease or the effects of aspirin intake and is able to measure the haemostatic effect of DDAVP. In our opinion, DDAVP should be administered only in cases of proved or clinically suspected platelet dysfunction, but clearly not as the first line for treating various bleeding disorders. As a matter of course bleeding due to thrombocytopaenia is treated immediately by transfusion of platelets.
If DDAVP, transfusion of fresh frozen plasma and platelets fail to stop bleeding, the administration of antifibrinolytic agents is recommended as the second step. In contrast to DDAVP, antifibrinolytic agents (aprotinin, tranexamic acid and aminocaproic acid) have been shown to reduce blood loss, especially in cardiac surgery, liver transplantation and some orthopaedic surgical procedures . Nevertheless, with regard to side-effects, antifibrinolytic agents should be given only if hyperfibrinolysis is a proven or suspected cause of the bleeding disorder. The detection of hyperfibrinolysis requires specific monitoring. Interestingly, the authors suggested the PFA-100® for detecting hyperfibrinolysis, although that device was designed and approved to detect platelet disorders, as already mentioned. However, thrombelastographic monitoring is able to detect hyperfibrinolysis and has been proven to be efficacious, especially in liver surgery and neurosurgery, where hyperfibrinolysis commonly occurs. If hyperfibrinolysis initiates diffuse microvascular bleeding, antifibrinolytics should be administered immediately and not after failed administration of DDAVP, fresh frozen plasma and platelets.
The authors correctly point out that a diffuse, non-surgical bleeding tendency is frequently caused initially by the combination of depletion and dilution of clotting factors. When blood loss is compensated for with colloids and red blood cell concentrates, deficiency of coagulation factors results, whereby fibrinogen deficiency appears first . Further, colloids are well known to disturb fibrin polymerization as well as the reticular fibrin network . The critical fibrinogen concentration of 75 mg dL−1 mentioned by the authors is definitively too low for surgical patients. Fibrinogen plasma levels below 1.5 g dL−1 and factor XIII plasma levels below 60% have been associated with an increased rate of post-operative bleeding complications . Additionally, fibrinogen plasma levels, as measured by the method described by Claus overestimate actual fibrinogen values in the presence of artificial colloid solutions. Although critical fibrinogen concentrations appear first, the authors omit the use of fibrinogen concentrates in their approach.
Concerning antithrombin, the large multicentre trial, mentioned by the authors, was published two years ago. High-dose antithrombin therapy had no effect on 28-day all-cause mortality in patients with severe sepsis and septic shock. Antithrombin administration was associated with increased risk for haemorrhage when administered with heparin . The recommendation that antithrombin be administered to minimize non-surgical bleeding is not reasonable. Antithrombin was suggested to diminish multiorgan failure and to improve survival in severely ill patients. In such patients diffuse microthrombus formation results in organ dysfunction, increased mortality and development of septic shock. However, antithrombin is an anticoagulant protein with no apparent effect on coagulation.
Contrary to the statement made by the authors, prothrombin-complex-concentrate does not contain factor V. A factor V deficiency can be treated only with fresh frozen plasma. Human recombinant activated factor XIII concentrates are not available, as purported by the authors. Factor XIII concentrates are derived from human plasma and are not activated.
In conclusion, the step-by-step approach presented by Menzebach and colleagues suggests a trial and error administration of various haemostatic agents without differentiation of the underlying pathology. For example, there is no distinction in diagnosis or therapy between congenital and acquired coagulopathy between dilutional and consumptive coagulopathy, platelet disorders, hyperfibrinolysis and disseminated intravascular coagulation. In our opinion, coagulation management perioperatively and on the intensive care unit implies coagulation monitoring and therapy. Point of care monitoring systems like thrombelastography and PFA-100® are very helpful for rapid diagnosis and initiation of appropriate treatment. These diagnostic measures need to be implemented by standard guidelines. If these systems are not available, a presumptive clinical diagnosis must be made. Again, we believe that the trial and error application of blood components, clotting factor concentrates or other haemostatic agents is ineffective and inappropriate.
the Perioperative Haemostasis
Task Force of the Austrian Society of Anaesthesia, Resuscitation and Intensive Care Medicine; Department of Anaesthesiology and Critical Care Medicine; Leopold-Franzens University; Innsbruck, Austria
1. Menzebach A, Cassens U, Van Aken H, Booke M. Strategies to reduce perioperative blood loss related to non-surgical bleeding. Eur J Anaesthesiol
2. Henry DA, Moxey AJ, Carless PA, et al.
Desmopressin for minimizing perioperative allogenic blood transfusion. Cochrane Database of Systematic Reviews
3. Porte RJ, Leebeek FW. Pharmacological strategies to decrease transfusion requirements in patients undergoing surgery. Drugs
4. Hiippala ST, Myllylä GJ, Vahtera EM. Hemostatic factors and replacement of major blood loss with plasma-poor red cell concentrates. Anesth Analg
5. Mardel SN, Saunders FM, Allen H, et al.
Reduced clot quality of clot formation with gelatin based plasma substitutes. Br J Anaesth
6. Gerlach R, Tölle F, Raabe A, et al.
Increased risk for postoperative hemorrhage after intracranial surgery in patients with decreased FXIII activity. Stroke
7. Warren BL, Eid A, Singer P, et al.
Caring for the critically ill patient. High-dose antithrombin III in severe sepsis: a randomized controlled trial. J Am Med Assoc