| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Departments of *Anaesthesiology and Critical Care Medicine, and General and Surgical Critical Care Medicine, Innsbruck Medical University, Innsbruck, Austria.
Address correspondence and reprint requests to Thorsten Haas, MD, Department of Anaesthesiology and Critical Care Medicine, Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria. Address e-mail to thorsten.haas{at}i-med.ac.at.
Abstract
BACKGROUND: During craniosynostosis repair, massive blood loss, consumption and dilution of clotting factors often result in coagulopathy, for which cryoprecipitate, fresh frozen plasma (FFP), and platelets are recommended for treatment. However, cryoprecipitate is not available in most European countries, and the efficacy of FFP in correcting fibrinogen deficiency is limited. We report our experience with human fibrinogen concentrate (Hemocomplettan®) used to improve impaired fibrinogen polymerization in children.
METHODS: Results of routine coagulation tests, thrombelastometry (ROTEM®), transfusion requirements, administration of fibrinogen concentrate, and data on the postoperative course of nine consecutive children undergoing major craniofacial surgery were retrospectively collected from anesthesia protocols, medical charts, laboratory and ROTEM® databases.
RESULTS: The nine children aged 12 (8, 22) mo (median [25th, 75th percentile]), weighing 9.5 (9, 10) kg had a calculated blood loss of 80 (49, 92)% of calculated blood volume during the surgery lasting 6.4 (4.5, 7.2) h. Impaired fibrinogen polymerization detected by ROTEM® was the main problem underlying dilutional coagulopathy. In all cases, sufficient hemostasis was achieved without adverse effects by administering (if necessary), repeated doses of fibrinogen concentrates (each single dose 30 mg/kg) without FFP or platelet transfusions. All children were successfully weaned from mechanical ventilation within a few hours and were able to be discharged early from the Intensive Care Unit.
CONCLUSIONS: Administration of fibrinogen concentrate effectively improves fibrinogen polymerization and total clot strength, which were the main underlying problems of dilutional coagulopathy in children undergoing craniosynostosis surgery.
Reconstructive craniofacial surgery for correction of craniosynostosis has optimal results when the operation is performed on younger patients. Unfortunately, these procedures carry a high risk of unavoidable and extensive blood loss during surgery, and every effort should be made to minimize allogeneic transfusion. Reported blood loss ranged between 20% and 500% of patient's estimated blood volume.1 A series of 1092 cases showed that hemorrhage was an important contributor to the perioperative mortality rate.2 Maintaining normovolemia by administering adequate amounts of fluids is essential. However, crystalloid and colloid intravascular volume replacement and transfusion of plasma-poor red cell concentrates can also cause dilutional coagulopathy. Thrombelastographic techniques (ROTEM®TEG®) usually show reduced clot strength as a first sign of dilutional coagulopathy, which results from impaired fibrinogen/fibrin polymerization, and might occur with moderate blood loss on the patient's initial fibrinogen concentrations.3–5 After a loss of 150%–200% of blood volume, significant dilution effects on other coagulation factors occur and thrombocytopenia usually develops as a cause of bleeding after transfusion of 15–20 units of blood.6–8 The current pediatric transfusion guidelines recommend the transfusion of cryoprecipitate when fibrinogen concentration is critically reduced (<100 mg/dL), fresh frozen plasma (FFP) should be transfused if standard coagulation tests are prolonged (more than 1.5 the normal value), and platelets are needed at counts <50,000–110,000/µL.9 However, evidence is growing that hemostasis management guided by bedside thrombelastographic techniques, which has been in routine use at our institution for several years, is associated with fewer allogeneic transfusions and might be preferable in the intraoperative situation.10–14 In addition, cryoprecipitate is not available in most European countries and FFP administration is inadequate for correcting fibrinogen deficiencies and exhibits unfavorable side effects. For several years, the administration of pasteurized human fibrinogen concentrates had been used in clinical practice when fibrinogen polymerization was the major underlying hemostasis disturbance, as detected by thrombelastographic techniques.12,15
We report our clinical experience in nine children where we used only fibrinogen concentrates (Haemocomplettan®, CSL Behring, Marburg, Germany) and red cell transfusions to treat blood loss and dilutional coagulopathy during craniosynostosis repair, thereby avoiding the need for FFP or platelet transfusion.
METHODS
From May 2005 to February 2006 nine children underwent major craniosynostosis repair (scaphocephalus [n = 2], plagiocephalus [n = 3], trigonocephalus [n = 3], turricephalus [n = 1]) at Innsbruck Medical University, Division of Maxillofacial Surgery. For each child, preoperative coagulation test results were available and no history or laboratory evidence of hereditary or acquired coagulopathy or thrombophilia was present.
Collected Data
We retrospectively collected from anesthesia charts, ROTEM® database, and medical charts the results of perioperative coagulation tests (prothrombin time [PT], activated partial thromboplastin time [aPTT], fibrinogen concentration), ROTEM® tracings (Pentapharm, Munich, Germany), blood cell count (hemoglobin [Hb], hematocrit, platelet count), data on intravascular volume supply, administration of pasteurized human fibrinogen concentrate, transfusion of red cells, FFP or platelets and postoperative course (cumulative drain output, revision surgery, time until tracheal extubation and discharge from intensive care unit). As the actual blood loss is usually underestimated, we also calculated perioperative blood loss using the Nadler and Mercuriali formulae.16,17
Anesthetic Management
In all children, anesthesia, fluid management, hemostatic therapy, and transfusion of blood components were performed according to clinical routine as described below. All children received midazolam orally for premedication (0.5 mg/kg), and anesthesia was induced using sevoflurane in oxygen/air mixture supplemented with remifentanil and rocuronium (0.6 mg/kg) to facilitate oral intubation. All children were kept normothermic and underwent invasive arterial and central venous monitoring. Fluid therapy was guided by central venous pressure (>7 cm H2O) and urine output (>1 · mL · kg–1 · h–1); dopamine (2–5 µg · kg–1 · min–1) was administered when fluid administration failed to maintain arterial blood pressure. Routine fluid management consisted of administration of 5% glucose in Ringer's lactated solution (RLG 10–20 mL · kg–1 · h–1) to cover deficiency in the fasting period and for basal demands. At the discretion of the anesthesiologist modified gelatin solution, 6% hydroxyethyl starch (HES) 130/0.4 or 5% albumin was additionally used to compensate blood loss (at a ratio of 1:1 to lost blood volume) until transfusion of red cells was necessary (Hb level <8 mg/dL). In all children, a cell salvage device was used to collect and process autologous blood (CATS®, Fresenius, Vienna, Austria).
Coagulation Monitoring and Therapy
Hemostatic therapy (administration of fibrinogen concentrate, FFP, or platelet transfusion) was directed by bedside ROTEM® analysis, although anesthesiologists had access to results of standard coagulation tests. Technical details of the ROTEM® analyzer have been described elsewhere.15 To analyze clot formation in citrated whole blood, we used the extrinsically (ExTEM®) and intrinsically activated (InTEM®) tests (activation with tissue factor or phospholipid-ellagic acid, respectively), and an extrinsically activated test containing the platelet-blocking substance cytochalasin D (FibTEM®) to separately evaluate functional fibrinogen polymerization. A typical ROTEM® tracing and its interpretation are given in Figure 1.
|
At our institution, based on our clinical experience, we routinely perform ROTEM® analysis in children as in adults, with an anticipated considerable blood loss (>30% of calculated blood volume) and intravascular volume demand at baseline (after induction of anesthesia) to monitor the hemostatic competence of the patient. In our children ROTEM® analyses were first repeated intraoperatively when the treating anesthesiologist observed increased microvascular bleeding from wound sites. Further analyses were performed to control the effects of hemostatic therapy or in any unclear situation in order to anticipate additional demands. Because the time of those analyses was dictated by the individual case, and thus varied considerably, we depicted and analyzed the baseline variables, which were obtained at comparable time points in all children, the first intraoperative variables before first fibrinogen administration, and those obtained at the end of surgery.
In the presence of microvascular bleeding (increased bleeding from surgical wound and catheter insertion sites), a fibrin clot strength (FibTEM® maximum clot firmness [MCF]) of 
7 mm (which is 25% less than the lower normal value), as analyzed by bedside ROTEM®, was the usual threshold for administering fibrinogen concentrate (Hemocomplettan®, CSL Behring, Marburg, Germany) at a single dose of 30 mg/kg via an infusion pump over 20 min. This therapy was repeated whenever microvascular bleeding and impaired fibrin polymerization were observed.
Furthermore, our concept allowed for administering FFP in cases with prolonged ROTEM® coagulation time (CT) and platelets when MCF remained <45 mm after correction of fibrinogen polymerization (FibTEM® MCF 
10 mm) or when the platelet count decreased to <50,000/µL. Postoperatively, all patients were transferred to the pediatric intensive care unit.
Statistics
Data were analyzed using SPSS (SPSS, Chicago, IL). A nonparametric Friedman ANOVA was used to detect overall differences between time points. In the case of a significant overall test, paired Wilcoxon Tests were performed for comparison between individual time points. Statistical significance was defined as P < 0.05.
RESULTS
All data are expressed as median (25th, 75th percentile). The nine children (ASA I-II) were at a median age of 12.0 (8, 22) months and a weight of 9.5 (9, 10) kg. Preoperatively the calculated total blood volume was 1077 (1046, 1136) mL. Children received RLG at amounts of 21 (14, 24) mL · kg–1 · h–1 throughout surgery, which lasted 6.4 (4.5, 7.2) h, and in eight of the nine children a dopamine infusion was administered. The calculated perioperative blood loss was 846 (514–1150) mL or 80 (49–92)% of calculated blood volume. Blood loss was compensated for by intermitted administration of colloids (gelatin solution, n = 3; HES 130/0.4, n = 4; 5% albumin, n = 2) at amounts of 360 (170, 870) mL, and all children required transfusion of salvaged and allogeneic red cells (200[100, 200] mL).
After 126 (82–146) min of surgery, values, as determined by standard laboratory tests (Hb, PT, aPTT, platelet count, and fibrinogen concentration), were significantly deteriorated as compared to baseline (Table 1) and all children showed impaired fibrin polymerization (FibTEM® MCF) less than the lower normal value, a decrease of about 15% in total clot strength (ExTEM® MCF) and a decrease in 
angle. At that time, children had received an infusion of 340 (220, 510) mL RLG and 150 (135, 200) mL colloid. CT detection, mainly a factor of concentrations of coagulation factors other than fibrinogen, remained unchanged. To restore fibrinogen polymerization and prevent a further reduction in total clot strength, a first dose of fibrinogen concentrate of 30 (25, 35) mg/kg was administered, targeted to reach a FibTEM® MCF of 8–10 mm and a total clot strength of >45 mm. Over the entire course of surgery, eight children required repetitive administration of fibrinogen (median 2 [2,3] times); the total perioperative dose of fibrinogen concentrate was 680 (600, 1500) mg, corresponding to 76 (67, 100) mg/kg. At the end of surgery, fibrinogen concentration, fibrin polymerization, and angle were significantly increased toward baseline as compared with the intraoperative values, and total clot strength remained within acceptable levels. A typical intraoperative ROTEM® series in one child is given in Figure 2. No child required transfusion of FFP or platelets during or after surgery.
|
|
During the first 24 h after surgery, median blood loss through surgical drains was minor at 60 (40, 95) mL. In all cases, no signs of excessive intravascular volume were present; weaning from mechanical ventilation was feasible within a few hours, and patients were discharged to the normal ward on the morning of the next day. Postoperative transfusion of red blood cells (68 mL) was necessary in only one patient. During their hospital stay no further transfusions were needed, no bleeding complications were noted and no signs of thromboembolic events were reported.
DISCUSSION
Our retrospectively collected data show that impaired fibrin polymerization and decreased plasma concentrations of fibrinogen are the primary problems encountered in children undergoing craniosynostosis repair and developing dilutional coagulopathy due to massive blood loss and intravascular volume demand. We also found that administering fibrinogen concentrates improves fibrinogen polymerization, thereby enabling total clot strength to be maintained without the need for platelet or FFP transfusion, even during continuing blood loss and further intravascular volume administration.
We used the ROTEM® device instead of conventional TEG®, which is also suited for bedside monitoring of hemostasis. However, because the technology of the two systems and the activators or platelet inhibitors we used show some differences and, moreover, reference values also differed slightly, this needs to be considered when interpreting our results.15,18
In addition, clinicians had access to results of standard coagulation tests, which, however, are poor predictors of increased bleeding,19 and were commonly available only after 45–60 min. In contrast, ROTEM®/TEG® monitoring enables a more timely and differential diagnosis of the underlying problem,12,15 thereby reducing the need for transfusion of blood components.20 We observed that PT and aPTT values soon became pathologic, thus reaching values of FFP transfusion triggers. As reported by Ciavarella et al.,21 such findings do not necessarily reflect the deficiency of coagulation factors, which are primarily needed for thrombin formation and initiation of coagulation. We found ROTEM® CT analysis of intrinsically and extrinsically activated tests (reflecting the concentration of enzymatic coagulation factors/inhibitors) unchanged and within normal values, even at the end of surgery, thus indicating adequate initiation of coagulation. Nevertheless, without ROTEM® monitoring and the concept of targeted therapy, the children would have at least received FFP, either on the basis of PT and aPTT results, as prophylaxis to prevent further dilution22 or even empirically in the presence of massive blood loss, and probably would have also received platelets.
Fibrinogen concentrates were administered when fibrin polymerization (FibTEM® MCF) decreased to less than the lower normal values and a clinically relevant increase in bleeding was observed. Thus, administration was not prophylactic. At this time, the measured concentrations of fibrinogen were in the range of 114–145 mg/dL (pediatric reference range 150–379 mg/dL), which is slightly higher than the threshold of <100 mg/dL commonly used in transfusion algorithms. The aim of our therapy was to maintain fibrin polymerization and not exact plasma fibrinogen levels. If no fibrinogen concentrate had been administered, total clot strength would have decreased still further and might have resulted in values indicative of platelet transfusion. A FibTEM® MCF of 7 mm corresponds to a fibrinogen concentration of about 150 mg/dL. However, this correlation is lost if HES starch is used, showing lower FibTEM® values at a given fibrinogen concentration.5 Hiippala et al.6 showed that the presence of colloids falsely increases fibrinogen concentrations, which are also difficult to standardize, especially in the lower and higher ranges.23
In fact, no clear critical threshold is known for fibrinogen concentration.9 In an in vitro study using TEG®, Nielsen et al.24 found that at fibrinogen levels <50 mg/dL no clot is formed, and at 75 mg/dL only weak clot formation occurs (–75% of normal). Confirming our and others'25,26 assumption that a threshold of 100 mg/dL fibrinogen is too low during continuing blood loss, a recently published study in parturients showed fibrinogen concentrations <200 mg/dL to have a 100% (71%–100%) predictive value for severity of postpartum hemorrhage.27 Considering all these findings, and our observations that microvascular bleeding starts at FibTEM® MCF 7 mm, we prefer to intraoperatively maintain fibrin polymerization at levels of 8–10 mm, which is still just at the lower normal value of 9 mm.
Because cryoprecipitate is not available at our institution or in most European countries, we used fibrinogen concentrate, which has been shown to be effective in treating dilutional coagulopathy in several in vitro, experimental, and observational studies10,28–31 and was used as long as impaired fibrinogen/fibrin polymerization was the only underlying problem. Because FFP and platelets produce undesirable side effects, such as transfusion-related acute lung injury and nonimmunologically triggered impairment of pulmonary function,32 prophylactic transfusion should be avoided. In addition, FFP administration is associated with remarkable intravascular volume expansion, and an increase in critically reduced concentrations of any protein of interest is difficult to achieve.30 Moreover, Miller et al.33 showed, in children undergoing cardiac surgery, that coagulation variables were more strongly impaired and that postoperative blood loss and need for transfusion of coagulation products was even greater in children receiving FFP after initial platelet transfusion as compared to those receiving platelets and fibrinogen by administration of cryoprecipitate.
Some limitations of the present observation need to be discussed. First, we report retrospectively collected data, thereby lacking a controlled design. For this reason, it could be assumed that the increase in fibrinogen concentration and the improved fibrinogen polymerization and cessation of bleeding occurred simply by chance. This is, however, very unlikely because during continuing blood loss fibrinogen synthesis cannot compensate for the increased consumption.34 In addition, microvascular bleeding and bleeding from spongy bone surfaces are hardly treatable by surgical means. Second, we cannot prove that with the presented regimen blood loss was reduced and associated with fewer transfusions of red cells as compared to a historical control group receiving no hemostatic therapy or FFP alone. We evaluated these data, but during the past 5 yr most of the children with major craniofacial procedures received FFP and additional fibrinogen concentrate. Analysis of data from children before the era of ROTEM® monitoring would have been influenced by surgical and anesthesiologic factors that have changed markedly over the last 10 yr. However, on a mathematical basis, at least 25 mL/kg FFP would have been required in our children instead of the administered fibrinogen concentrate.
Third, no reference values for ROTEM® parameters in small children have been confirmed by investigating large cohorts of children. Thus, our regimen relied on reference values reported for adults.35 Principally, the vitamin K-dependent coagulation factors, as well as anticoagulatory proteins, are present at birth at about 50% of adult values and gradually increase to about 80% during the first 6 mo, whereas plasma concentrations of fibrinogen, coagulation factors V, VIII, XIII, and von Willebrand factor are not decreased during infancy.36 In an investigation on the influence of colloids on hemostasis in 42 children, we observed that FibTEM® MCF values and plasma fibrinogen concentration showed good correlation, even in children <1 yr,37 and the baseline ROTEM® parameters were within the range reported for adults, although frequently at a borderline level. Lastly, one may argue that coagulopathy was iatrogenic due to intravascular volume overload or hypervolemic hemodilution. Because red cells were transfused at a median of only 200 mL, but blood loss was about 800 mL, the administered amount of crystalloids/colloids was reasonable.
In conclusion, during pediatric craniosynostosis repair impairment of fibrinogen polymerization and decreased total clot strength developed as a result of massive blood loss and necessary intravascular volume therapy and was successfully treated by substituting only human fibrinogen concentrate, thereby preventing the need for FFP or platelets. Our experience needs to be confirmed by controlled studies, which, however, are difficult to conduct, because FFP and platelets influence coagulation, as does fibrinogen concentrate, although differently. In addition, in pediatric and adult studies showing considerable blood loss, and thus the development of coagulopathy, the inclusion of a control group receiving no hemostatic therapy is impossible for ethical reasons.
Footnotes
Accepted for publication November 20, 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  N. Rahe-Meyer, M. Pichlmaier, A. Haverich, C. Solomon, M. Winterhalter, S. Piepenbrock, and K. A. Tanaka Bleeding management with fibrinogen concentrate targeting a high-normal plasma fibrinogen level: a pilot study Br. J. Anaesth., June 1, 2009; 102(6): 785 - 792. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
