Amniotic fluid (AF) accelerates clot formation in in vitro conventional coagulation tests of blood mixed with AF (1–4). The end point of these tests is the first appearance of a clot, the time of which was shortened by AF. Thrombelastograph® (TEG; Haemoscope Corporation, Skokie, IL) can provide more information, including the initiation, propagation, strength, and lysis of a clot (5). TEG is increasingly used to assess coagulation and guide treatment in cardiac and liver surgery (6–8). Such guidance can reduce transfusion of blood products, especially of fresh frozen plasma (7,9). Sharma et al. (10), using TEG, showed that blood from pregnant women is hypercoagulable compared with blood from nonpregnant women, with accelerated clot formation and increased clot strength. AF embolism can cause complex changes in coagulation, ranging from enhanced coagulation to fibrinolysis (11), and TEG testing may be useful in its management. Our aim was to demonstrate both the enhanced coagulation and fibrinolytic effects of AF on normal blood in a single in vitro test, using TEG.
Institutional ethics committee approval and consent from patients and obstetricians were obtained. AF was obtained from 10 healthy patients having amniocentesis in midtrimester and from 10 healthy patients during elective caesarean delivery at full term. In the caesarean delivery patients, the AF was drawn after dissection of the uterus down to the amniotic membrane. Samples with visible contamination by blood were not used. Women with preeclampsia were excluded. The AF was centrifuged at 1000 rpm for 5 min to remove any cellular elements. Only the supernatant was used in the TEG tests.
Venous blood was obtained from 10 healthy nonpregnant female volunteers immediately before commencing the TEG tests. As pregnancy can cause a hypercoagulable state (10), we used blood from nonpregnant volunteers as we wanted to assess the effect of both midtrimester and final-trimester AF on the same normal blood. Volunteers who had been taking any nonsteroidal antiinflammatory drugs were excluded. A two-syringe technique was used. The first 2 mL of sampled blood was discarded to minimize contamination of the blood by tissue activators. Four computerized TEG machines (Haemoscope Corporation) were used for the study.
Mallet and Cox (5) reviewed the TEG devices. The main working parts of the TEG are a heated cup, which is oscillated, and a pin that is suspended freely from a torsion wire. Blood (0.36 mL) is placed in the cup. Before the formation of clot, the movement of the cup does not affect the pin. Once the clot starts to form, the pin is connected to the cup by fibrin strands and moves with the cup. The shear modulus and elasticity of the clot is transmitted through the pin and is displayed as the TEG trace with the following variables: the reaction time (R), the time from the placement of the blood in the cup to the first sign of clot formation; the time from R to a fixed level of clot firmness (K) (trace amplitude reaches 20 mm), which reflects fibrin buildup and cross-linking; the angle (α) formed by the slope of the TEG tracing between the R and K points, which reflects the rate of clot growth; the maximal amplitude (MA), the greatest amplitude that the TEG trace achieves, which reflects platelet quantity and function and clot strength; and the percent lysis at 30 (LY30) and 60 (LY60) min after MA, which reflect the extent of fibrinolysis. The coagulation index (CI) is a computer calculated linear combination of the R, K, MA, and α measurements. It reflects overall coagulation of the sample. The normal range for nonpregnant subjects is −3.0 to +3.0 (12). A more positive result indicates hypercoagulability and a more negative result indicates hypocoagulability.
The TEG machines were prewarmed to 37°C. Midterm AF (5, 10, 20 μL) and similar volumes of full-term AF were placed in six cuvettes in the TEG machines. Venous blood from a volunteer was then added into the cuvettes to produce a total volume of 360 μL. The seventh cuvette only had 360 μL of blood and was used as the control sample for both mid- and full-term AF. The TEG tests were run simultaneously and the R, K, α, MA, LY30, and LY60 variables recorded.
All TEG variables were compared by using analysis of variance, and Bonferroni’s test was used for multiple comparisons (SPSS 8.0 for Windows, SPSS Inc., Chicago, IL). Paired t-tests were used to determine statistical significance of differences between full-term and midterm AF. A P value of < 0.05 was con- sidered significant in all tests. All data were reported as mean ± SD.
All control TEG tests of the blood from volunteers were within normal limits. A representative normal nonpregnant whole blood control TEG trace and the TEG trace showing enhanced coagulation and fibrinolysis after adding 20 μL of final-trimester AF are shown in Figure 1.
The gestational age was 20 ± 1 wk. The TEG measurements are shown in Table 1 and in Figure 2A. With increasing volumes of AF, there was a significant shortening of the R (P < 0.001) and K times (P < 0.05) when compared with the control sample. There was also a significant (P < 0.05) increase in the α at all volumes of AF. There were no significant changes in the MA, LY30, and LY60 with the addition of midtrimester AF.
Final-Trimester (Full-Term) AF
The gestational age of the pregnancies was 39 ± 1 wk. The TEG measurements are also shown in Table 1 and Figure 2B. The tests showed enhanced coagulation, with significant and progressive shortening of R (P < 0.001) and K (P < 0.05) times with increasing volumes of AF. The α also increased significantly as increasing volumes of AF were added (P < 0.05). There were no significant changes in the MA, LY30, and LY60 with the addition of final-trimester AF.
The CI was significantly increased at all volumes of midtrimester and final-trimester AF added compared with the controls (P < 0.05), again reflecting enhanced coagulation (Table 1 and Figure 2). However, all the CI results were within the normal range for blood from nonpregnant subjects.
In comparing the TEG results of midtrimester and final-trimester AF, there were no significant overall differences, although isolated comparisons showed statistical significance. The K time was shorter, and α and MA were greater with 10 μL of final-trimester AF compared with 10 μL of midtrimester AF. There were no significant differences in R, LY30, LY60, and CI between midtrimester and final-trimester AF at any of the volumes used.
In this study using TEG analysis, we have shown that AF significantly accelerates coagulation but does not significantly increase fibrinolysis of native whole blood. Both initiation and propagation of clot formation are faster, as shown in the significant shortening of R and K times, and the significant increase in the α and CI with increasing volumes of AF. However, even though clot formation was accelerated, we did not demonstrate increased clot strength as reflected by the nonsignificant changes in MA.
The mechanism of accelerated coagulation appears to be multifactorial. Weiner et al. (1) showed that thromboplastin released from fetal cells in AF promotes the conversion of prothrombin to thrombin, and fibrinogen to fibrin. Phillips and Davidson (2) found that AF could shorten the clotting times of both normal plasma and plasma deficient in factors VII, VIII, IX, and XI. AF, in in vitro tests, not only resembled thromboplastin in activating the extrinsic pathway but also involved the intrinsic pathway. They asserted that amniotic fluid contained a factor that could activate factor X and quantified this in terms of Russell’s viper venom equivalents. This effect of AF may have a role in primary hemostasis in the uterus after delivery.
We were unable to show significantly increased fibrinolysis with the addition of AF, even though there was a trend toward increased lysis with higher volumes of full-term AF. Gross dissolution of the clots did not occur. AF contains a tissue plasminogen activator, urokinase-like plasmin activator and thrombin antithrombin complexes, which can cause the enhanced fibrinolysis seen (13). The enhanced coagulation caused by AF may also contribute to increased activation of fibrinolytic system. Increased fibrinolysis has been shown to occur in AF embolism (14).
The effects of AF on coagulation have been related to gestational age (15). By using only the supernatant, which is free of cellular elements, both mid and final-trimester AF showed similar effects in our study. There were no significant differences between midtrimester and final-trimester AF. Further study with more patients stratified for gestational age is required.
AF embolism is a rare complication, with an incidence of between 1 in 20,000 to 1 in 80,000 pregnancies, but with mortality as high as 80% (16,17). Complex effects on coagulation have been reported in the 40% of patients who do not die from cardiovascular collapse (11,14,17). There is initially a thromboplastic effect and disseminated intravascular coagulation. Fibrinolysis, hypofibrinogenemia, and a hemorrhagic state follow. A hemorrhagic tendency is the main clinical effect seen. Whether this is caused by defective coagulation or increased clot lysis may not be elucidated by conventional tests, which only reflect isolated portions of the hemostatic system. Even though this study did not show increased fibrinolysis, TEG can guide rational treatment with blood products as it can give more complete information in each patient.
While TEG variables may not always correlate well with conventional tests of coagulation (such as prothrombin time, activated partial thromboplastin time, and platelet count), especially in hypocoagulable patients, this is because TEG measures different processes (6,18,19). TEG assesses the interaction of clotting factors and cellular elements, which is not reflected in tests of isolated portions of the coagulation cascade.
In summary, we have demonstrated in in vitro TEG tests that AF accelerates clot initiation and propagation of native whole blood. The global coagulation picture of TEG analysis may be useful in assessing coagulopathy in patients with AF embolism.
We would like to thank Eli Cohen, PhD, for his comments. We would also like to thank Karen Ho of the Department of Anesthesia, and the staff of the Department of Obstetrics and Gynaecology, National University Hospital, Singapore, for their help.
1. Weiner AE, Reid DE, Roby CC. The hemostatic activity of amniotic fluid. Science 1949; 110:190–1.
2. Phillips LL, Davidson EC. Procoagulant properties of amniotic fluid. Am J Obstet Gynecol 1972; 113:911–9.
3. Lockwood CJ, Bach R, Guha A, et al. Amniotic fluid contains tissue factor, a potent initiator of coagulation. Am J Obstet Gynecol 1991; 165:1335–41.
4. Yaffe H, Hay-Am E, Sadovsky E. Thromboplastic activity of amniotic fluid in term and postmature gestations. Obstet Gynecol 1981; 57:490–2.
5. Mallet SV, Cox DJA. Thrombelastography. Br J Anaesth 1992; 69:307–13.
6. Tuman KJ, McCarthy RJ, Ivankovich AD. The thrombo-elastograph: is it the solution to coagulation problems? Cardiothoracic and vascular anesthesia update. Philadelphia: WB Saunders, 1991: 1–13.
7. Spiess BD, Gillies BSA, Chandler W, Verrier E. Changes in transfusion therapy and re-exploration rate after institution of blood management program in cardiac surgical patients. J Cardiothorac Vasc Anesth 1995; 9:168–73.
8. Kang Y. Coagulation and liver transplantation. Transplant Proc 1993; 25:2001–5.
9. Essell JH, Martin TJ, Salinas J, et al. Comparison of thromboelastography to bleeding time and standard coagulation tests in patients after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 1993; 7:410–5.
10. Sharma SK, Philip J, Wiley J. Thrombelastographic changes in healthy parturients and postpartum women. Anesth Analg 1997; 85:94–8.
11. Stolte LAM, Lim HT, Van Arkel C, et al. Amniotic fluid embolism. In: Fairwater DVI, Eskes TKAB, eds. Amniotic fluid: research and clinical application. Amsterdam: Encerpta Medica, 1973: 313–32.
12. Thrombelastograph® coagulation analyzer model 3000 series: user manual—theory and operation. Skokie, IL: Haemoscope Corporation, 1998:6.
13. Koh SCL, Anandakumar C, Arulkumaran S, et al. Amniotic fluid plasminogen activators and inhibitors and thrombin anti-thrombin complex levels during second trimester and labour. Fibrinolysis 1995; 9:121–6.
14. Ratnoff OD, Vosburgh GH. Observations on the clotting defect in amniotic fluid embolism [letter]. N Engl J Med 1952; 247:970.
15. Hastwell GB. Accelerated clotting time: an amniotic fluid thromboplastic activity index of fetal maturity. Am J Obstet Gynecol 1978; 131:650–4.
16. Clark SL, Montz, FJ, Phelan JP. Hemodynamic alterations associated with amniotic fluid embolism: a reappraisal. Am J Obstet Gynecol 1985; 151:617–20.
17. Clark SL. New concepts of amniotic fluid embolism: a review. Obstet Gynecol Surv 1990; 45:360–8.
18. Zuckerman L, Cohen E, Vagher JP, et al. Comparison of thromboelastography with common coagulation tests. Thromb Haemosts 1981; 46:752–6.
© 2000 International Anesthesia Research Society
19. Tuman KJ, Speiss BD, McCarthy RJ, Ivankovich AD. Comparison of viscoelastic measures of coagulation after cardiopulmonary bypass. Anesth Analg 1989; 69:69–75.