Secondary Logo

Journal Logo

Pediatric Anesthesia: Society for Pediatric Anesthesia

Predicting and Treating Coagulopathies After Cardiopulmonary Bypass in Children

Miller, Bruce E. MD; Mochizuki, Toshiaki MD; Levy, Jerrold H. MD; Bailey, James M. MD, PhD; Tosone, Steven R. MD; Tam, Vincent K. H. MD; Kanter, Kirk R. MD

Author Information
  • Free


Bleeding after cardiopulmonary bypass (CPB) is a major cause of hemodynamic instability and morbidity after cardiac surgery. Whereas investigations in adults have cited acquired platelet function abnormalities as the major etiologic factor of bleeding [1-4], the coagulopathies after CPB in pediatric patients may be much more complex because of age- and disease-related factors. Although correlations have been found between certain coagulation tests after CPB and the amount of subsequent chest tube drainage in adults [5,6], these correlations have not been adequately studied in the pediatric population. Therefore, the transfusion of component coagulation factors to restore hemostasis after protamine administration is empiric. Although one study has shown the usefulness of fresh whole blood in correcting coagulopathies and limiting post-CPB blood loss in children less than two years of age, [7] the availability of whole blood is limited in many pediatric cardiac centers. In light of the limited information in pediatric patients, our objectives were to define demographic factors and/or coagulation tests that correlated with postoperative chest tube drainage and to evaluate the effectiveness of different coagulation products (platelets, cryoprecipitate, and fresh frozen plasma) in correcting pertinent coagulation tests and restoring hemostasis after CPB.


After approval by our human investigations committee and informed parental consent, the following information was collected in 76 pediatric patients of all ages and weights who were not taking drugs known to interfere with coagulation and whose surgery required CPB and was scheduled as the first case of the day. Before surgical incision, prothrombin time (PT), activated partial thromboplastin time (aPTT), platelet count, fibrinogen level, activated clotting time (ACT) (Hemochron system; International Technidyne Corporation, Edison, NJ), and Thrombelastograph[registered sign] (TEG[trademark symbol]) tracing (Haemoscope Corporation, Skokie, IL) were obtained. TEG values measured included R, K, alpha angle, maximal amplitude (MA), and amplitude 60 min after MA was reached (A-60). The fibrinolytic index (A-60/MA) was then calculated to define fibrinolysis (A-60/MA<0.85). Coagulation index (CI = -[0.1227]R + [0.0092]K + [0.1665]MA - [0.0241] alpha - 5.0020) (Thrombelastograph[registered sign] Operations Manual and CTEG User's Guide, Haemoscope Corp.) and shear modulus (G = [5000]MA/100 - MA) [8] were also calculated, as these have been shown to be more sensitive measures of coagulation and actual clot strength than individual TEG parameters. After the administration of 400 U/kg body weight of bovine heparin (Upjohn, Kalamazoo, MI) via a central venous catheter and documentation of ACT values exceeding 480 s, CPB was conducted with a membrane oxygenator (Cobe VPCML Plus; Cobe Cardiovascular, Inc., Arvada, CO) at temperatures determined by the complexity of the planned surgical procedure. The volume of the pump prime was based on the patient's weight: 750 mL for patients less than 7 kg, 950 mL for patients between 7 and 22 kg, and 1200 mL for patients larger than 22 kg. Packed red blood cells were added to maintain an acceptable hematocrit during CPB. ACT values were maintained in excess of 480 s by the administration of additional heparin if needed. Aprotinin was not administered to any patient.

After CPB, the protamine dose was determined by heparin-protamine titrations. Ten minutes after protamine administration and with the ACT at baseline, platelet counts, fibrinogen levels, and TEG values were again obtained. The decision to administer component therapy was then made jointly by the anesthesia and surgical teams based only on the appearance of the surgical field because of the relatively long turnover time required to obtain coagulation tests and because, on conception of this study, no known correlations had been defined between laboratory tests and post-CPB bleeding in children. Platelet transfusions were first given to any patient in whom bleeding was considered to be excessive after protamine administration. In the event of continued bleeding after platelet transfusion, cryoprecipitate or fresh-frozen plasma (FFP) was administered, with the choice, in part, depending on the availability of donor-directed products. ACT and TEG measurements were repeated in the operating room after the administration of each individual coagulation product. On arrival in the intensive care unit (ICU), repeat PT, aPTT, platelet counts, and fibrinogen levels were obtained. The amount of chest tube drainage during the first 24 h postoperatively was recorded, as was the necessity for the administration of additional coagulation factors during this period.

A stepwise regression analysis was used to determine demographic factors, baseline coagulation tests, or CPB data that would be predictive of 24-h chest tube drainage. Weight and duration of CPB were found to be the only predictors. Because weight is reflective of several pathophysiologic factors that could influence the coagulation system and because it is a variable that is known preoperatively, we further analyzed this variable using two-sided t-tests to compare weights at 1-kg intervals to define the weight below which chest tube drainage was significantly greater. Once this weight was defined, two-sided t-tests were also used to compare baseline and postprotamine coagulation values of patients above and below this weight. The Pearson product-moment correlation coefficient was calculated to determine whether there was a significant relationship between individual coagulation tests and postoperative chest tube drainage in each group. Analysis of variance was used to determine whether there were differences in coagulation values among subgroups when analyzing the effects of coagulation product transfusions. Comparisons of the subgroups were then made using two-sided t-tests, assuming unequal variances with Bonferroni's correction for multiple comparisons.


One patient died 8 h postoperatively of complications unrelated to bleeding and therefore was excluded from subsequent analyses. No patient required reexploration for excessive bleeding. After determining the correlation between weight and 24-h chest tube drainage by regression analysis and by using two-sided t-tests to evaluate chest tube drainage at 1-kg weight intervals, children weighing <8 kg (n = 43) were found to have significantly more 24-h chest tube drainage than children >8 kg (n = 32) (38.8 +/- 23.6 vs 25.8 +/- 17.5 mL/kg; P = 0.008). Demographic data differed between these two weight groups. (Table 1). Baseline PT, aPTT, and TEG R and K values were significantly longer in patients weighing <8 kg. Although duration of CPB, another predictor of postoperative chest tube drainage, was not different between these weight groups, postprotamine platelet count, fibrinogen level, ACT, and TEG MA, CI, and G values demonstrated a more severe coagulation defect in the smaller patients after CPB (Table 2). Two patients weighing >8 kg, but none of the smaller patients, demonstrated TEG evidence of fibrinolysis after protamine administration. One of these patients had met the TEG criteria for fibrinolysis preoperatively. Transfusion of coagulation products was avoided in 8 of 32 patients weighing >8 kg (25%) but only in 1 of 43 patients weighing <8 kg (2%).

Table 1
Table 1:
Patient Demographics
Table 2
Table 2:
Coagulation Test Comparisons Between the <8 kg and >8 kg Groups

No baseline coagulation test was found by regression analysis to be predictive or, by correlation coefficient, to correlate with postoperative chest tube drainage. After protamine administration to patients weighing <8 kg, platelet count and fibrinogen level correlated independently with eventual chest tube drainage. In addition, both platelet count and fibrinogen level correlated with TEG alpha, MA, and G values at this point, although these TEG values were not found to correlate with chest tube drainage in this group. After protamine administration to patients weighing >8 kg, platelet count, as well as TEG alpha and MA parameters, correlated independently with chest tube drainage. Platelet count again correlated with TEG MA and G values, whereas fibrinogen level correlated with no TEG values, nor with chest tube drainage, in this group.

We examined each of the two weight groups individually to determine the hemostatic effects of the transfusion of different coagulation products. In each weight group, four patient subgroups were established based on the products administered in the operating room (OR): no coagulation products, platelets only, platelets followed by cryoprecipitate, or platelets followed by FFP. Three patients received a combination of platelets, cryoprecipitate, and FFP (two weighing <8 kg and one weighing >8 kg) and are excluded from further comment. Only two patients weighing <8 kg received no coagulation products in the OR. One of these patients subsequently received platelets in the 24-h postoperative study period. Because of this small number, this subgroup was also excluded from further comparisons. However, 10 patients weighing >8 kg received no coagulation products in the operating room. Of these 10 patients, 8 received no products at all during the 24-h postoperative study period, so this subgroup was included in further analyses.

In the group weighing <8 kg, 17 received only platelet transfusions in the OR, 12 received platelets followed by cryoprecipitate, and 10 received platelets followed by FFP. Demographic data, baseline coagulation tests, and the duration of CPB were the same among these subgroups. After protamine administration, platelet count and fibrinogen level, as well as TEG alpha, MA, CI, and G values, remained significantly different from baseline values for the entire group. The administration of platelets to all patients improved these variables but returned only the TEG alpha and CI values to baseline levels. In patients who ultimately received only platelets, however, the TEG MA value was also restored to baseline by the platelet transfusion alone. The additional transfusion of cryoprecipitate to patients with continued bleeding returned the fibrinogen level and TEG MA and G values to baseline. The administration of FFP to treat ongoing bleeding after platelet transfusion not only did not significantly improve fibrinogen level but also worsened all TEG values (Table 3). The subgroup of patients requiring only platelet transfusions had the least amount of 24-h chest tube drainage (34.5 +/- 16.2 mL/kg), and patients subsequently receiving cryoprecipitate accumulated less chest tube drainage than did the patients receiving FFP, although these differences did not reach statistical significance (38.8 +/- 18.7 vs 44.2 +/- 30.1 mL/kg). During the first 24 h postoperatively, 7 of 17 patients (41%) who received only platelets in the OR, 7 of 12 patients (58%) in the platelet-cryoprecipitate subgroup, and 7 of 10 patients (70%) in the platelet-FFP subgroup were given additional coagulation products in the ICU.

Table 3
Table 3:
Coagulation Test in Patients Weighing <8 kg

In the group weighing >8 kg, 10 patients received no coagulation products in the OR. A comparison of these patients with all patients who were transfused coagulation products in the OR revealed no significant difference in weight. For the 21 patients who later received coagulation products, the duration of CPB was longer (128 +/- 60 vs 66 +/- 38 min; P = 0.001), postprotamine platelet counts were lower (86 +/- 28 vs 111 +/- 25; P = 0.02), and postprotamine TEG values revealed a significantly greater coagulopathy. Of these 21 patients, 7 received only platelets, 5 received platelets followed by cryoprecipitate, and 9 received platelets followed by FFP. Demographic data and baseline coagulation values were the same among these three subgroups. Patients in the platelet-FFP subgroup were exposed to CPB longer than those receiving no products or only platelets, but this duration was not different from the platelet-cryoprecipitate subgroup. After protamine administration, platelet count and TEG alpha and MA values, as well as fibrinogen level and TEG R, CI, and G values, remained significantly different from baseline values for the entire group. Transfusion of platelets to all patients in whom continued bleeding was deemed to be excessive restored only the TEG R value to baseline, despite improving all of these coagulation parameters. However, in patients who ultimately received only platelets, TEG alpha and MA values were also returned to levels not significantly different from baseline by the platelet transfusion alone. The subsequent administration of cryoprecipitate to patients with ongoing bleeding after platelet transfusion returned not only fibrinogen level but also TEG MA, CI, and G values to baseline. The administration of FFP to combat continued bleeding after platelet transfusion once again resulted in the deterioration of all coagulation parameters (Table 4). Additionally, only the 24-h chest tube drainage of the platelet-FFP subgroup was significantly greater than that of patients who received no coagulation products. Patients in the subgroups receiving only platelets or platelets and cryoprecipitate had the same amount of 24-h chest tube drainage (22.5 +/- 9.6 or 20.2 +/- 10.2 mL/kg, respectively), and both of these subgroups had remarkably less drainage than did the platelet-FFP subgroup (39.8 +/- 19.4 mL/kg). During the 24-h postoperative study period, 2 of 10 patients (20%) who received no coagulation products in the OR, 0 of 7 (0%) who received only platelets, 1 of 5 (20%) who received platelets and cryoprecipitate, and 5 of 9 (56%) who received platelets and FFP were given additional coagulation products in the ICU.

Table 4
Table 4:
Coagulation Tests in Patients Weighin >8 kg


Our investigation determined the weight and duration of CPB to be predictive of 24-hour chest tube drainage after CPB in children. We found no baseline coagulation test to be predictive. Because weight is reflective of other factors influencing the coagulation system (age-related immaturity, decompensation from congestive heart failure, degree of hemodilution on CPB), is known preoperatively, and could be used to anticipate a patient's post-CPB coagulation status, we further analyzed our patients by weight. We found that children smaller than 8 kg have significantly more chest tube drainage per body weight and require transfusion of more coagulation products during the first 24 hours after CPB than do larger patients. Postprotamine platelet count and fibrinogen level were observed to correlate independently with chest tube drainage in patients weighing <8 kg, as were postprotamine platelet count and TEG alpha and MA values in patients weighing >8 kg. Although platelet transfusion was found to improve coagulation and attenuate bleeding to acceptable levels in many patients in both weight groups, maximal improvement was gained when continued bleeding was treated with cryoprecipitate administration after platelet transfusion. Patients receiving FFP after platelet transfusion subsequently had worsened coagulation parameters, more 24-hour chest tube drainage, and more transfusions of coagulation products in the ICU compared with the other subgroups.

Increased postoperative chest tube drainage in small children may be related to the more severe post-CPB coagulopathies that these patients acquire. Our smaller patients had more baseline coagulation abnormalities, most likely resulting from the quantitative factor deficiencies of infancy [9,10] and the coagulation disturbances that accompany cyanosis and congestive heart failure [11,12]. These abnormalities are exacerbated by the profound hemodilution produced on exposure to the proportionately larger priming volumes of the CPB circuit [13]. The significant differences in coagulation tests after protamine between our weight groups demonstrate the more severe post-CPB coagulopathy of smaller children (Table 2). The fact that only 1 of 43 patients weighing less than 8 kg versus 8 of 32 patients weighing more than 8 kg received no coagulation products during the study period adds indirect support for this observation.

No preoperative coagulation test was useful in predicting post-CPB chest tube drainage in these children. Argument could therefore be made about the economic futility of requiring these tests in preparation for surgery. Only quantitative tests (platelet count and fibrinogen level) correlated with postoperative chest tube drainage in the smaller patients after protamine, whereas functional TEG data also became useful in the larger patients. Because minimal levels of the coagulation proteins are necessary to allow for coagulation [1], extreme deficiencies in the smaller patients after CPB caused marked abnormalities in qualitative tests, thereby rendering these tests useless in predicting chest tube drainage. With less extreme quantitative factor deficiencies in the larger patients, qualitative tests (TEGs) became more indicative of the patient's coagulation status and therefore correlated with chest tube drainage. We retrospectively found TEG data to be an indicator of the need to administer any coagulation products at all after protamine or to transfuse additional products after platelets, although we administered coagulation products based only on clinical judgment in this study.

The transfusion of platelets to children in either weight group with continued bleeding after protamine administration improved all coagulation parameters and was sufficient to restore adequate hemostasis in many cases. Because children acquire platelet function abnormalities after CPB in the same manner as adults [7,14] and because platelet count is greatly reduced by hemodilution during CPB [13,14], the initial transfusion of platelets is logical. In situations of continued bleeding after platelet administration, the transfusion of cryoprecipitate rather than FFP restored coagulation values to baseline in both groups. No benefit in correcting abnormal coagulation values or restoring hemostasis was found by the transfusion of FFP.

Cryoprecipitate contains concentrated amounts of fibrinogen, factor VIII, von Willebrand factor (vWF), and factor XIII. Fibrinogen's adherence to the platelet aggregation receptor, glycoprotein IIb/IIIa, and vWF's interactions with the platelet adhesive receptor, glycoprotein Ib, are important for maximizing platelet function. Because the number of both of these receptors is decreased after CPB, increasing fibrinogen and vWF levels may enhance the function of the remaining receptors [4,15-17]. Additionally, restoring levels of factor XIII, which is responsible for cross-linking fibrin monomers and whose post-CPB deficiency has been shown in adults to decrease clot strength and TEG MA values and to correlate with blood loss, should also improve hemostasis [18,19]. The benefit and advantage of cryoprecipitate is its ability to replenish fibrinogen, vWF, and factor XIII levels with a small amount of transfused volume. Each unit of cryoprecipitate consists of only 5-15 mL and contains essentially all of the fibrinogen and up to 70% of the vWF and 30% of the factor XIII found in a 250-mL unit of FFP [20]. We transfused approximately one-half unit of cryoprecipitate per kilogram body weight to our smaller patients. Administration of enough FFP to restore these important factor levels would impose too large a volume load to these small patients, thus limiting the usefulness of FFP in correcting coagulopathies after CPB in children.

The current investigation would have benefited by assigning matched patients to specific subgroups and transfusing only the designated coagulation products for continued bleeding. However, the availability of directed-donor FFP influenced the choice of coagulation products. Despite this limitation, we feel that our comparison of the effects of cryoprecipitate versus FFP is important information. Similar numbers of patients in each group received cryoprecipitate and FFP. No differences were found in demographics; coagulation tests at baseline, after protamine, or after platelet transfusion; or duration of CPB between these two subgroups in patients weighing less than or more than 8 kg. Therefore, the cryoprecipitate and FFP subgroups contained similar patients with coagulopathies of similar severity.

In summary, by using regression analysis, we found weight and duration of CPB to be predictors of post-CPB chest tube drainage in children. No preoperative coagulation test is useful in this regard and, therefore, the economic impact of obtaining these tests should be considered. Further analyses involving weight showed 8 kg to be a critical weight, below which post-CPB coagulopathies should be expected to be more severe, 24-hour chest tube drainage higher, and transfusion requirements greater. Only quantitative tests (platelet count and fibrinogen level) correlate independently with postoperative chest tube drainage in patients less than 8 kg, whereas functional tests (TEG alpha and MA) also have significant correlations in larger children. TEG data are also useful in evaluating the effectiveness of transfused products in all children. Platelet transfusion restores adequate hemostasis in many children, but in the presence of continued bleeding, cryoprecipitate is remarkably more effective than FFP in correcting coagulopathies and reducing chest tube drainage.


1. Harker LA, Malpass TW, Branson HE, et al. Mechanism of abnormal bleeding in patients undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with selective alpha-granule release. Blood 1980;56:824-34.
2. Mammen EF, Koets MH, Washington BC, et al. Hemostasis changes during cardiopulmonary bypass surgery. Semin Thromb Hemost 1985;11:281-92.
3. Zilla P, Fasol R, Groscurth P, et al. Blood platelets in cardiopulmonary bypass operations: recovery occurs after initial stimulation, rather than continual activation. J Thorac Cardiovasc Surg 1989;97:379-88.
4. George JN, Pickett EB, Saucerman S, et al. Platelet surface glycoproteins: studies on resting and activated platelets and platelet membrane microparticles in normal subjects, and observations in patients during adult respiratory distress syndrome and cardiac surgery. J Clin Invest 1986;78:340-8.
5. Spiess BD, Tuman KJ, McCarthy RJ, et al. Thromboelastography as an indicator of post-cardiopulmonary bypass coagulopathies. J Clin Monit 1987;3:25-30.
6. Tuman KJ, Spiess BD, McCarthy RJ, Ivankovich AD. Comparison of viscoelastic measures of coagulation after cardiopulmonary bypass. Anesth Analg 1989;69:69-75.
7. Manno CS, Hedberg KW, Kim HC, et al. Comparison of the hemostatic effects of fresh whole blood, stored whole blood, and components after open heart surgery in children. Blood 1991;77:930-6.
8. Speiss BD, et al. Anesthesiology 1994;81:A566.
9. Andrew M, Paes B, Milner R, et al. Development of the human coagulation system in the full-term infant. Blood 1987;70:165-72.
10. Andrew M, Paes B, Milner R, et al. Development of the human coagulation system in the healthy premature infant. Blood 1988;72:1651-7.
11. Kontras SB, Sirak HD, Newton WA. Hematologic abnormalities in children with congenital heart disease. JAMA 1966;195:99-103.
12. Colon-Otero G, Gilchrist GS, Holcomb GR, et al. Preoperative evaluation of hemostasis in patients with congenital heart disease. Mayo Clin Proc 1987;62:379-85.
13. Kern FH, Morana NJ, Sears JJ, Hickey PR. Coagulation defects in neonates during cardiopulmonary bypass. Ann Thorac Surg 1992;54:541-6.
14. Chan AKC, Leaker M, Burrows FA, et al. Coagulation and fibrinolytic profile of paediatric patients undergoing cardiopulmonary bypass. Thromb Haemost 1997;77:270-7.
15. Musial J, Niewiarowski S, Hershock D, et al. Loss of fibrinogen receptors from the platelet surface during simulated extracorporeal circulation. J Lab Clin Med 1985;105:514-22.
16. Sakariassen KS, Bolhuis PA, Sixma JJ. Human blood platelet adhesion to artery subendothelium is mediated by factor VIII-Von Willebrand factor bound to the subendothelium. Nature 1979;279:636-8.
17. Huang H, Ding W, Su Z, Zhang W. Mechanism of the preserving effect of aprotinin on platelet function and its use in cardiac surgery. J Thorac Cardiovasc Surg 1993;106:11-8.
18. Shainoff JR, Estafanous FG, Yared JP, et al. Low factor XIIIa levels are associated with increased blood loss after coronary artery bypass grafting. J Thorac Cardiovasc Surg 1994;108:437-45.
19. Girolami A, de Marco L, Dal Bo Zanon R, et al. Rarer quantitative and qualitative abnormalities of coagulation. Clin Haematol 1985;14:385-411.
20. Stehling LC. Blood component therapy. In: Lake CL, Moore RA, eds. Blood: hemostasis, transfusion, and alternatives in the perioperative period. New York: Raven Press, 1995:277-88.
© 1997 International Anesthesia Research Society