Arch Gynecol Obstet DOI 10.1007/s00404-015-3768-0
MATERNAL-FETAL MEDICINE
Anticoagulant management of pregnant women with mechanical heart valve replacement during perioperative period Ce Bian1 • Xiaorong Qi1 • Li Li1 • Jitong Zhao1 • Xinghui Liu1
Received: 23 November 2014 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Objective To investigate the morbidity of complications and pregnancy outcomes in women with mechanical heart valve replacement who received low-dose oral anticoagulation treatment with warfarin throughout the pregnancy, compare the prognosis and complications of patients who were treated with single oral warfarin treatment or the ‘‘bridging’’ therapy treatment, investigate the influence of using vitamin K1 before emergency cesarean section delivery on postoperative warfarin anticoagulant effect and to explore an appropriate anticoagulant regimen during perioperative period for pregnant women with mechanical heart valve replacement. Method 46 pregnant women with mechanical heart valve replacement who received low-dose oral anticoagulation treatment from October 2008 to October 2014 treated at West China Women’s and Children’s Hospital were retrospectively reviewed. Eight patients received emergency cesarean section (CS), while 38 patients received selective CS, in which 17 patients received single oral warfarin and 21 patients received ‘‘bridging’’ anticoagulation treatment during postoperative period. Morbidity of complications and the time to achieve the target INR after operation were compared. Results The mechanical valves were at the mitral position in 35 (76.09 %) patients, at the aortic position in 2 & Xinghui Liu [email protected] 1
Department of Obstetrics and Gynecology, Sichuan Provincial Key Laboratory of Gynecologic Oncology, Key Laboratory of Obstetric and Gynecologic and Pediatric Diseases and Birth Defects of Ministry of Education, West China Women’s and Children’s Hospital, Sichuan University, Chengdu 610041, Sichuan, People’s Republic of China
(4.35 %) patient and at both the mitral and aortic position in 9 (19.57 %) patients. 46 full-term healthy babies were delivered and no maternal thromboembolic was observed during pregnancy. There was no significant difference of the amount of uterine bleeding between single oral warfarin group and ‘‘bridging’’ treatment group during postpartum period. In single oral warfarin group, one valve thrombosis was observed and led to sudden death. No periphery thrombosis, hematoma, general hemorrhage or other sign of over-anticoagulation was observed. The INR increased more slowly in the group who received emergency CS with preoperative application of vitamin K1 than other two groups. Conclusion The use of vitamin K1 preoperatively might result in warfarin resistance and discontinuation of warfarin therapy before selective CS might be more appropriate than application of vitamin K1. The ‘‘bridging’’ anticoagulation treatment which combines oral warfarin and subcutaneous LMWH might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement during postoperative period, no matter selective or emergency CS. The safety of low-dose oral warfarin therapy throughout pregnancy is still under controversy. Keywords Mechanical heart valve Pregnancy Anticoagulation Warfarin Vitamin K1 Perioperative period
Introduction Pregnancy prompts hypercoagulability state, which increases the risk of mechanical heart valve thrombosis and death [1], and effective anticoagulation should be
123
Arch Gynecol Obstet
mandatory [2]. Oral anticoagulation offers the best protection for mothers, with a lower incidence of valve thrombosis than unfractionated heparin (UFH) and lowmolecular-weight heparin (LMWH), whose application, on the other hand, is associated with fetal malformation and pregnancy loss [3]. Coumarin derivatives cross the placenta and their use in the first trimester can result in embryopathy in 0.6–10 % of cases [4–7]. UFH and LMWH do not cross the placenta and embryopathy does not occur. But more evidence shows that the hypercoagulable risk of coumarin derivatives may be dose dependent [6, 8]. Therefore, lowdose oral anticoagulation therapy throughout pregnancy may be a simple, safe and more acceptable regimen. Because of the discontinuation of warfarin, the hypercoagulable state, and patients presenting no or sluggish increase of INR during postpartum warfarin resumption, patients are more susceptible to mechanical valve thrombosis during the perioperative period. Therefore, the regimen during this period must be more effective. The purpose of this study was to retrospectively investigate the morbidity of complications in women with mechanical heart valve replacement who received lowdose oral anticoagulation treatment with warfarin throughout the pregnancy, compare the prognosis and complications of patients who were treated with single oral warfarin treatment or the ‘‘bridging’’ anticoagulation treatment during perioperative period, investigate the influence of using vitamin K1 before emergency cesarean section (CS) on postoperative warfarin anticoagulant effect and to explore an appropriate anticoagulant regimen during perioperative period for pregnant women with mechanical heart valve replacement.
Materials and methods We conducted a retrospective review of all patients identified with mechanical heart valve replacement in West China Women’s and Children’s Hospital from October 2008 to October 2014. Eligible patients in this analysis were women received low-dose oral anticoagulation therapy throughout pregnancy and terminated pregnancy by CS. Women with spontaneous delivery, abortion and incomplete clinical data were ineligible. Out of the population, 46 consecutive patients with mechanical heart valve replacement were identified from the obstetric database. Clinical information regarding preoperative characteristics including age, reproductive history, underlying heart disease, position of the mechanical valve, age at valve replacement and gestational weeks were obtained from patients’ records. Written informed consent was obtained from all participants and this study was approved by the Institutional Ethics Committee of Sichuan University.
123
Patients were advised to contact the outpatient clinic as soon as they miss a period, and to perform pregnancy tests every 3 days until positive or menstruation. Upon confirmation of pregnancy, every pregnant woman should be informed that oral anticoagulation offers the best protection against thrombosis, but is associated with an appreciable risk of fetal malformations and pregnancy loss. On the other hand, substitution of oral anticoagulation with LMWH or UFH reduces the risk of fetal damage, but increases the risk of valve thrombosis, even when administered in adjusted doses. Each pregnant patient completed written informed consent. The international normalized ratio (INR) was estimated on a weekly basis at our outpatient clinic and recorded along with prescribed warfarin doses to maintain the target INR between 1.6 and 2.2. Echocardiographic follow-up was performed monthly to evaluate cardiac and prosthetic function. Patients were followed up by cardiologists and obstetricians at biweekly intervals until the 36th week of gestation, when they were estimated weekly. Selective CS was typically planned at 38 weeks of gestation. Warfarin therapy was discontinued 2 days before surgery. During the preoperative period, LMWH was not routinely administered, as was collegially considered not necessary for a warfarin discontinuation within just 3 days [9], and INR was checked daily. Echocardiography was done before the operation. The maternal thrombotic, hemorrhagic complications and pregnancy outcomes were investigated from patients’ database. During postoperative period, patients of selective cesarean section received single oral warfarin or the ‘‘bridging’’ anticoagulation treatment. Single oral warfarin group reinitiated warfarin (2.5 mg/day), 4–12 h after delivery, depending on the degree of uterine bleeding; and the ‘‘bridging’’ treatment group reinitiated warfarin, and administered subcutaneous LMWH twice daily to maintain anti-Xa levels of 1–1.2 IU/ml until the target INR was reached. The uterine bleeding of the two groups during 24, 24–48, 48–72 h after surgery and the morbidity of postoperative complications such as maternal valve thrombosis, periphery thrombosis, general hemorrhage and other overanticoagulation were compared. For the patients who needed emergency cesarean section delivery, vitamin K1 (10 mg) was given intravenously, and INR was checked every 2–4 h until INR reached normal. Depending on the degree of uterine bleeding, ‘‘bridging’’ treatment was administered 4–12 h after delivery and the INR level was examined daily. The time to achieve the target INR after the emergency and selective cesarean section was compared. Statistical analysis Statistical analysis was performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). Uterine bleeding and time
Arch Gynecol Obstet
for anticoagulation therapy to CS were compared using one-way ANOVA and serial changes in INR levels were compared by RM-ANOVA. p value \ 0.05 was considered significant.
Results From October 2008 to October 2014, 46 patients with mechanical heart valve replacement received low-dose oral anticoagulation therapy throughout pregnancy and terminated pregnancy by CS in West China Women’s and Children’s Hospital. The baseline characteristics of the patients are shown in Table 1. All patients were primipara with a mean age of 30.54 years old (range 20–42) and age at valve replacement was 25.04 years old (range 12–39). The termination time of pregnancy was 38.33 weeks (range 37.1–40.3). In these 46 patients, 44 (95.65 %) cases suffered from rheumatic heart disease while the other 2 (4.35 %) cases suffered from congenital heart diseases. The mechanical valves were at the mitral position in 35 (76.09 %) patients, at the aortic position in 2 (4.35 %) patient and at both the mitral and aortic position in 9 (19.57 %) patients. Valve replacement was conducted 5.48 years (range 1–17) before pregnancy, and warfarin anticoagulation was routinely used after the surgery. All patients received low-dose oral anticoagulation therapy throughout pregnancy and were regularly monitored coagulation functions during Table 1 Baseline characteristics of patients with mechanical heart valve replacement Characteristics
(n = 46)
Maternal age (years) Mean
30.54
Range
20–42
Age at valve replacement (years) Mean Range Primigravida (n, %)
25.04 12–39 46 (100 %)
Gestational age (weeks) Mean
38.33
Range
37.1–40.3
Underlying heart disease (n, %) Rheumatic
44 (95.65 %)
Congenital
2 (4.35 %)
Position of the mechanical valve (n, %) Mitral
35 (76.09 %)
Aortic
2 (4.35 %)
Mitral ? Aortic
9 (19.57 %)
pregnancy and regularly followed up by cardiologists. The usage of warfarin was all less than 5 mg/day and daily dose was 1.25–4.375 mg, the INR was controlled between 1.6 and 2.2. No maternal thromboembolic was observed during pregnancy. Sixteen patients (34.78 %) suffered from general bleeding, in which subcutaneous ecchymosis occurred in nine patients (19.57 %), epistaxis occurred in five patients (10.87 %), and gingival bleeding occurred in two patents (4.35 %). Heart failure occurred during CS in one patent (2.17 %), but recovered after emergency treatment. Other cardiac complications occurred in 15 patients (32.61 %), in which 7 patients (15.22 %) suffered from atrial fibrillation and 8 patients (17.39 %) suffered from ventricular premature beat. 46 full-term babies were born and the mean birth weight was 2978.65 g (range 2180–4000). In these 46 neonates, there were two low birth weight infants, their weight were 2180 g and 2290 g, respectively, and were transferred to department of neonatology after birth. Apgar scores of 46 neonates at birth were more than 7 points, with no neonatal or fetal death, no stillbirth, no congenital abnormalities or neurological dysfunctions. No warfarin embryopathy, characterized by nasal hypoplasia, stippled epiphyses, or both, was observed. Thirty-eight patients received selective CS, in which 17 patients received single oral anticoagulation treatment and 21 patients received ‘‘bridging’’ anticoagulation treatment during postoperative period. There was one maternal thromboembolic event in the oral therapy group and led to death. The patient was a 21-year-old woman with a history of rheumatic heart disease and received mitral valve replacement at the age of 12. She was treated with warfarin 3.75 mg/day. Her INR level ranged from 1.6 to 2.2 throughout pregnancy and received a selective CS at the 38?2 weeks of gestation. Echocardiography before the operation demonstrated no valve thrombosis but pulmonary artery hypertension. The operation went smoothly, and the warfarin was reverted 8 h after the operation. On the third day after the operation, the patient felt chest discomfort, and in the meanwhile, her INR level was 0.89. Thereafter, she acutely became short of breath, cardiac arrest and died soon. The autopsy demonstrated mitral mechanical valve thrombosis of approximately 16 mm 9 14 mm. The time for reinitiating anticoagulation therapy after CS was 7.65 h (range 4–11) in single oral warfarin group and 7.48 h (range 4–11) in ‘‘bridging’’ treatment group, where no statistically significant difference was found between two groups (p = 0.961). There was no significant difference in the amount of uterus bleeding during 3 days after surgery between two groups. No late postpartum hemorrhage, periphery thrombosis, general hemorrhage or any other over-anticoagulation complication were observed
123
Arch Gynecol Obstet Table 2 The comparisons between single oral warfarin group and ‘‘Bridging’’ treatment group Group
n
Time for Therapy to CS (h)
Uterine Bleeding (ml) 24 h
24–48 h
48–72 h
Valve Thrombosis
Periphery Thrombosis
Bridging
21
7.48
241.67
147.24
83.10
0
0
Oral
17
7.65
235.00
140.59
79.41
1
0
p
0.787
0.661
in two groups. The comparisons between two groups are shown in Table 2. Due to premature rupture of membranes and patients refused to choose natural childbirth, eight patients received emergency cesarean Section. 10 mg vitamin k1 was given intravenously, and operation was underwent until INR reached normal. ‘‘Bridging’’ treatment was administered 7.50 h (range 4–10) after delivery and the INR level was examined daily. Serial changes in INR levels during postpartum period were compared between vitamin K1 treatment group, single oral warfarin group and ‘‘bridging’’ treatment group. There was no significant difference in time to reinitiated anticoagulation therapy between these three groups. The INR levels in vitamin K1 treatment group increased more slowly than other two groups, which need 6.38 ± 1.06 days to achieve the target range. The ‘‘bridging’’ treatment group and oral anticoagulation group needed 6.42 ± 0.81 days and 4.71 ± 0.92 days, respectively. Significant difference is shown between ‘‘vitamin K1 treatment’’ group and the other two groups (p \ 0.05), with no significant difference between the other two groups (p = 0.767). The comparison of serial changes in INR levels between three groups is shown in Fig. 1.
Discussion Pregnancy induces a series of haemostatic changes, with an increase in concentration of coagulation factors, fibrinogen, and platelet adhesiveness, as well as diminished fibrinolysis, which lead to hypercoagulability and an increased risk of thromboembolic events. In addition, obstruction to venous return by the enlarging uterus causes stasis and a further rise in risk of thromboembolism. Therefore, anticoagulation therapy is important for pregnant women who had received mechanical heart valve replacement. To avoid intraoperative and postoperative hemorrhage, the coagulation function of the patients should be maintained normal. In addition, the high maternal complication rate including valve thrombosis and resultant death should not be ignored either. Therefore, an effective anticoagulation therapy, after the bleeding tendency, is particularly important. Since the mid 1980s, heart surgeon has recognized that excessive
123
0.699
0.561
Fig. 1 The comparison of serial changes in INR between single oral warfarin group, ‘‘bridging’’ treatment group and vitamin K1 treatment group
anticoagulation strength is the important reason of postoperative bleeding and reduced the anticoagulation strength. The value of PTR has dropped to 1.5–2.0 from 3.0 to 4.0 and INR was dropped to 2.0–3.0 from 4.0 to 6.0. In this strength, the rate of bleeding was 1.4–2.4 % and the embolism was 2.0–3.8 % in western countries [10]. But in the same standard of anticoagulant strength, Chinese patients showed higher bleeding rate (0.68–10.4 %) and lower embolism rate (0.3–1.48 %) [11]. This shows that because of the difference between the races, it is necessary to use low intensity of anticoagulant in Chinese patients. And some studies also showed low-intensity anticoagulation was safe and effective for Chinese patients [8, 12]. So, the patients in our study received oral low-dose warfarin anticoagulation throughout pregnancy, with INR controlled between 1.6 and 2.2, and no complications happened, such as embolism and severe bleeding. It shows that the oral anticoagulant treatment of warfarin for pregnant women is safe and effective.
Arch Gynecol Obstet
In our study, we observed that even though the warfarin was reinitiated as soon as possible, the INR level still needed 4.71 ± 0.92 days to reach the target range. Patients who discontinued warfarin preoperatively were in the ‘unprotecting’ state, which were subjected to high risk of mechanical valve thrombosis until their postoperative INR reached the target range. In the single oral warfarin group, there was one patient suffered from mitral valve thrombosis and resulted in death. Therefore, single oral warfarin therapy might be insufficient for anticoagulation during perioperative period. LMWH works fast, and it takes only 3 h after subcutaneous injection that the peak concentration in serum is reached with almost 100 % bioavailability. Moreover, it is easier to achieve stable anticoagulation and evaluation. Moreover, LMWH is reported to have lower incidence of osteoporosis and thrombocytopenia compared with unfractionated heparin [13]. The ‘‘bridging’’ anticoagulation treatment reinitiates warfarin, and administers subcutaneous LMWH twice daily to maintain anti-Xa levels of 1–1.2 IU/ml until the target INR is reached. In our study, one out of 17 patients without postoperative ‘‘bridging’’ therapy died due to valve thrombosis and instead of increasing the morbidity of postpartum hemorrhage, late postpartum hemorrhage and general hemorrhage, the ‘‘bridging’’ treatment effectively prevented the occurrence of mechanical valve thrombosis, periphery thrombosis and stoke. Therefore, ‘‘bridging’’ treatment might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement after selective CS. Since LMWH does not influence APTT, not only INR but also anti-Xa levels should be checked daily during postoperative period. Because the anticoagulant effect of warfarin can be antagonized by vitamin K1, it is a common way to use high dose of vitamin K1 to correct coagulant disturbance when postoperative patients who received valve replacement are in over-anticoagulant state or before emergency operation. However, high doses of vitamin K1 might induce ‘‘warfarin resistance’’ [14], which could influence anticoagulant effect of warfarin even during postoperative period. In our study, it is obvious that the postoperative increase of INR in the vitamin K1 treatment group was significant more slow than other two groups. Hence, for the selective CS, the warfarin therapy should be discontinued 2 days before surgery instead of vitamin K1 application via intravenous infusion. For the emergency CS, due to the use of vitamin K1 and ‘‘warfarin resistance’’, the ‘‘bridging’’ treatment might be more appropriate and effective than single oral warfarin therapy during postoperative period until the target INR is reached. Warfarin provides the most effective control of anticoagulation with mechanical heart valves. In addition, it is considered to be the best method to prevent maternal
thrombotic events [15]. However, the best prophylaxis becomes questionable during pregnancy, for being unsafe to the fetus. Warfarin could cross the placenta and affect the fetus, which lead to ‘‘warfarin embryopathy’’, miscarriage and stillbirth [16]. Since 1991, Cotrufo et al. supported that those complications could be dose related, and suggested that warfarin would be safe as long as its use during pregnancy was limited to 5 mg/day [17]. However, the safety of low-dose oral warfarin therapy during first trimester is still under controversy [18]. In our study warfarin was continuously used throughout pregnancy and no warfarin embryopathy or over-anticoagulation in the fetus was observed, but since the objects and including criteria, this result might not reflect the real impact of warfarin on the fetus. However, there are several limitations to our study that must be considered. First, this is a retrospective, singlecenter study with a limited ability to reliably determine heart function and complications during pregnancy by patients’ records. Second, due to the limitations of knowledge and conditions, we did not use prothrombin complex concentrates but vitamin K1 to normalize the INR values prior to emergency cesarean section, which might increase the risk of patients. Third, we did not follow-up the neonatal outcomes, such as nervous system development, mental development, so this study could not reflect the impact of long-term development of fetus. Forth, the overall patient population is certainly limited, and the subgroup undergoing ‘‘vitamin K1 treatment’’ is even smaller. These limitations might be solved by increasing number of cases and participation of more research centers in future, and our study results could be confirmed by the larger sample study. And the use of prothrombin complex concentrates prior to emergency cesarean section could improve the safety of pregnant women with mechanical heart valve replacement. The discontinuation of warfarin during perioperative period increases the risk of mechanical valve thrombosis and death. The use of vitamin K1 preoperatively might result in warfarin resistance and discontinue warfarin therapy before selective CS might be more appropriate than application of vitamin K1. The ‘‘bridging’’ anticoagulation treatment which combines oral warfarin and subcutaneous LMWH might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement during postoperative period, no matter selective or emergency CS. The safety of low-dose oral warfarin therapy throughout pregnancy is still under controversy, the role of us is to continue the discussion of the pros and cons of available regimens and, ultimately, help the patient and her family to decide which drug to take. We consider that the most important is the mother’s life, second is the baby’s life or avoiding embryopathy.
123
Arch Gynecol Obstet Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Limet R, Grondin CM (1977) Cardiac valve prostheses, anticoagulation, and pregnancy. Ann Thorac Surg 23(4):337–341 2. Salazar E, Izaguirre R, Verdejo J, Mutchinick O (1996) Failure of adjusted doses of subcutaneous heparin to prevent thromboembolic phenomena in pregnant patients with mechanical cardiac valve prostheses. J Am Coll Cardiol 27(7):1698–1703 3. Pieper PG, Balci A, Van Dijk AP (2008) Pregnancy in women with prosthetic heart valves. Neth Heart J 16(12):406–411 4. Chan WS, Anand S, Ginsberg JS (2000) Anticoagulation of pregnant women with mechanical heart valves: a systematic review of the literature. Arch Intern Med 160(2):191–196 5. Schaefer C, Hannemann D, Meister R, Elefant E, Paulus W, Vial T, Reuvers M, Robert-Gnansia E, Arnon J, De Santis M, Clementi M, Rodriguez-Pinilla E, Dolivo A, Merlob P (2006) Vitamin K antagonists and pregnancy outcome. A multi-centre prospective study. Thromb Haemost 95(6):949–957 6. Vitale N, De Feo M, De Santo LS, Pollice A, Tedesco N, Cotrufo M (1999) Dose-dependent fetal complications of warfarin in pregnant women with mechanical heart valves. J Am Coll Cardiol 33(6):1637–1641 7. van Driel D, Wesseling J, Sauer PJ, Touwen BC, van der Veer E, Heymans HS (2002) Teratogen update: fetal effects after in utero exposure to coumarins overview of cases, follow-up findings, and pathogenesis. Teratology 66(3):127–140 8. Bian C, Wei Q, Liu X (2012) Influence of heart-valve replacement of warfarin anticoagulant therapy on perinatal outcomes. Arch Gynecol Obstet 285(2):347–351 9. De Santo LS, Romano G, Della Corte A, D’Oria V, Nappi G, Giordano S, Cotrufo M, De Feo M (2012) Mechanical aortic
123
10.
11.
12.
13.
14. 15.
16.
17.
18.
valve replacement in young women planning on pregnancy: maternal and fetal outcomes under low oral anticoagulation, a pilot observational study on a comprehensive pre-operative counseling protocol. J Am Coll Cardiol 59(12):1110–1115 Cannegieter SC, Torn M, Rosendaal FR (1999) Oral anticoagulant treatment in patients with mechanical heart valves: how to reduce the risk of thromboembolic and bleeding complications. J Intern Med 245(4):369–374 Dong L, Shi YK (1996) Anticoagulation therapy of patients with mechanical heart valve prostheses: follow-up survey of 126 cases. Chin J Thorac Cardiovasc Surg 12:149–151 Dong L, Shi YK, Tian ZP, Ma JY, Wang X, Yi J (2003) Low intensity anticoagulation therapy after mechanical heart valve replacement. Zhonghua wai ke za zhi [Chin J Surg] 41(4):250–252 Montalescot G, Polle V, Collet JP, Leprince P, Bellanger A, Gandjbakhch I, Thomas D (2000) Low molecular weight heparin after mechanical heart valve replacement. Circulation 101(10):1083–1086 Patriquin C, Crowther M (2011) Treatment of warfarin-associated coagulopathy with vitamin K. Expert Rev Hematol 4(6):657–665 Higa T, Okura H, Tanemoto K, Yoshida K (2014) Prosthetic valve thrombosis caused by heparin-induced thrombocytopenia thrombosis during pregnancy. Circ J 78(4):1004–1005 Cade JF, Hirsh J, Martin M (1969) Placental barrier to coagulation factors: its relevance to the coagulation defect at birth and to haemorrhage in the newborn. Br Med J 2(5652):281–283 Cotrufo M, de Luca TS, Calabro R, Mastrogiovanni G, Lama D (1991) Coumarin anticoagulation during pregnancy in patients with mechanical valve prostheses. Eur J Cardiothorac Surg 5(6):300–304 Hassouna A, Allam H (2014) Limited dose warfarin throughout pregnancy in patients with mechanical heart valve prosthesis: a meta-analysis. Interact Cardiovasc Thorac Surg 18(6):797–806
MATERNAL-FETAL MEDICINE
Anticoagulant management of pregnant women with mechanical heart valve replacement during perioperative period Ce Bian1 • Xiaorong Qi1 • Li Li1 • Jitong Zhao1 • Xinghui Liu1
Received: 23 November 2014 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Objective To investigate the morbidity of complications and pregnancy outcomes in women with mechanical heart valve replacement who received low-dose oral anticoagulation treatment with warfarin throughout the pregnancy, compare the prognosis and complications of patients who were treated with single oral warfarin treatment or the ‘‘bridging’’ therapy treatment, investigate the influence of using vitamin K1 before emergency cesarean section delivery on postoperative warfarin anticoagulant effect and to explore an appropriate anticoagulant regimen during perioperative period for pregnant women with mechanical heart valve replacement. Method 46 pregnant women with mechanical heart valve replacement who received low-dose oral anticoagulation treatment from October 2008 to October 2014 treated at West China Women’s and Children’s Hospital were retrospectively reviewed. Eight patients received emergency cesarean section (CS), while 38 patients received selective CS, in which 17 patients received single oral warfarin and 21 patients received ‘‘bridging’’ anticoagulation treatment during postoperative period. Morbidity of complications and the time to achieve the target INR after operation were compared. Results The mechanical valves were at the mitral position in 35 (76.09 %) patients, at the aortic position in 2 & Xinghui Liu [email protected] 1
Department of Obstetrics and Gynecology, Sichuan Provincial Key Laboratory of Gynecologic Oncology, Key Laboratory of Obstetric and Gynecologic and Pediatric Diseases and Birth Defects of Ministry of Education, West China Women’s and Children’s Hospital, Sichuan University, Chengdu 610041, Sichuan, People’s Republic of China
(4.35 %) patient and at both the mitral and aortic position in 9 (19.57 %) patients. 46 full-term healthy babies were delivered and no maternal thromboembolic was observed during pregnancy. There was no significant difference of the amount of uterine bleeding between single oral warfarin group and ‘‘bridging’’ treatment group during postpartum period. In single oral warfarin group, one valve thrombosis was observed and led to sudden death. No periphery thrombosis, hematoma, general hemorrhage or other sign of over-anticoagulation was observed. The INR increased more slowly in the group who received emergency CS with preoperative application of vitamin K1 than other two groups. Conclusion The use of vitamin K1 preoperatively might result in warfarin resistance and discontinuation of warfarin therapy before selective CS might be more appropriate than application of vitamin K1. The ‘‘bridging’’ anticoagulation treatment which combines oral warfarin and subcutaneous LMWH might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement during postoperative period, no matter selective or emergency CS. The safety of low-dose oral warfarin therapy throughout pregnancy is still under controversy. Keywords Mechanical heart valve Pregnancy Anticoagulation Warfarin Vitamin K1 Perioperative period
Introduction Pregnancy prompts hypercoagulability state, which increases the risk of mechanical heart valve thrombosis and death [1], and effective anticoagulation should be
123
Arch Gynecol Obstet
mandatory [2]. Oral anticoagulation offers the best protection for mothers, with a lower incidence of valve thrombosis than unfractionated heparin (UFH) and lowmolecular-weight heparin (LMWH), whose application, on the other hand, is associated with fetal malformation and pregnancy loss [3]. Coumarin derivatives cross the placenta and their use in the first trimester can result in embryopathy in 0.6–10 % of cases [4–7]. UFH and LMWH do not cross the placenta and embryopathy does not occur. But more evidence shows that the hypercoagulable risk of coumarin derivatives may be dose dependent [6, 8]. Therefore, lowdose oral anticoagulation therapy throughout pregnancy may be a simple, safe and more acceptable regimen. Because of the discontinuation of warfarin, the hypercoagulable state, and patients presenting no or sluggish increase of INR during postpartum warfarin resumption, patients are more susceptible to mechanical valve thrombosis during the perioperative period. Therefore, the regimen during this period must be more effective. The purpose of this study was to retrospectively investigate the morbidity of complications in women with mechanical heart valve replacement who received lowdose oral anticoagulation treatment with warfarin throughout the pregnancy, compare the prognosis and complications of patients who were treated with single oral warfarin treatment or the ‘‘bridging’’ anticoagulation treatment during perioperative period, investigate the influence of using vitamin K1 before emergency cesarean section (CS) on postoperative warfarin anticoagulant effect and to explore an appropriate anticoagulant regimen during perioperative period for pregnant women with mechanical heart valve replacement.
Materials and methods We conducted a retrospective review of all patients identified with mechanical heart valve replacement in West China Women’s and Children’s Hospital from October 2008 to October 2014. Eligible patients in this analysis were women received low-dose oral anticoagulation therapy throughout pregnancy and terminated pregnancy by CS. Women with spontaneous delivery, abortion and incomplete clinical data were ineligible. Out of the population, 46 consecutive patients with mechanical heart valve replacement were identified from the obstetric database. Clinical information regarding preoperative characteristics including age, reproductive history, underlying heart disease, position of the mechanical valve, age at valve replacement and gestational weeks were obtained from patients’ records. Written informed consent was obtained from all participants and this study was approved by the Institutional Ethics Committee of Sichuan University.
123
Patients were advised to contact the outpatient clinic as soon as they miss a period, and to perform pregnancy tests every 3 days until positive or menstruation. Upon confirmation of pregnancy, every pregnant woman should be informed that oral anticoagulation offers the best protection against thrombosis, but is associated with an appreciable risk of fetal malformations and pregnancy loss. On the other hand, substitution of oral anticoagulation with LMWH or UFH reduces the risk of fetal damage, but increases the risk of valve thrombosis, even when administered in adjusted doses. Each pregnant patient completed written informed consent. The international normalized ratio (INR) was estimated on a weekly basis at our outpatient clinic and recorded along with prescribed warfarin doses to maintain the target INR between 1.6 and 2.2. Echocardiographic follow-up was performed monthly to evaluate cardiac and prosthetic function. Patients were followed up by cardiologists and obstetricians at biweekly intervals until the 36th week of gestation, when they were estimated weekly. Selective CS was typically planned at 38 weeks of gestation. Warfarin therapy was discontinued 2 days before surgery. During the preoperative period, LMWH was not routinely administered, as was collegially considered not necessary for a warfarin discontinuation within just 3 days [9], and INR was checked daily. Echocardiography was done before the operation. The maternal thrombotic, hemorrhagic complications and pregnancy outcomes were investigated from patients’ database. During postoperative period, patients of selective cesarean section received single oral warfarin or the ‘‘bridging’’ anticoagulation treatment. Single oral warfarin group reinitiated warfarin (2.5 mg/day), 4–12 h after delivery, depending on the degree of uterine bleeding; and the ‘‘bridging’’ treatment group reinitiated warfarin, and administered subcutaneous LMWH twice daily to maintain anti-Xa levels of 1–1.2 IU/ml until the target INR was reached. The uterine bleeding of the two groups during 24, 24–48, 48–72 h after surgery and the morbidity of postoperative complications such as maternal valve thrombosis, periphery thrombosis, general hemorrhage and other overanticoagulation were compared. For the patients who needed emergency cesarean section delivery, vitamin K1 (10 mg) was given intravenously, and INR was checked every 2–4 h until INR reached normal. Depending on the degree of uterine bleeding, ‘‘bridging’’ treatment was administered 4–12 h after delivery and the INR level was examined daily. The time to achieve the target INR after the emergency and selective cesarean section was compared. Statistical analysis Statistical analysis was performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). Uterine bleeding and time
Arch Gynecol Obstet
for anticoagulation therapy to CS were compared using one-way ANOVA and serial changes in INR levels were compared by RM-ANOVA. p value \ 0.05 was considered significant.
Results From October 2008 to October 2014, 46 patients with mechanical heart valve replacement received low-dose oral anticoagulation therapy throughout pregnancy and terminated pregnancy by CS in West China Women’s and Children’s Hospital. The baseline characteristics of the patients are shown in Table 1. All patients were primipara with a mean age of 30.54 years old (range 20–42) and age at valve replacement was 25.04 years old (range 12–39). The termination time of pregnancy was 38.33 weeks (range 37.1–40.3). In these 46 patients, 44 (95.65 %) cases suffered from rheumatic heart disease while the other 2 (4.35 %) cases suffered from congenital heart diseases. The mechanical valves were at the mitral position in 35 (76.09 %) patients, at the aortic position in 2 (4.35 %) patient and at both the mitral and aortic position in 9 (19.57 %) patients. Valve replacement was conducted 5.48 years (range 1–17) before pregnancy, and warfarin anticoagulation was routinely used after the surgery. All patients received low-dose oral anticoagulation therapy throughout pregnancy and were regularly monitored coagulation functions during Table 1 Baseline characteristics of patients with mechanical heart valve replacement Characteristics
(n = 46)
Maternal age (years) Mean
30.54
Range
20–42
Age at valve replacement (years) Mean Range Primigravida (n, %)
25.04 12–39 46 (100 %)
Gestational age (weeks) Mean
38.33
Range
37.1–40.3
Underlying heart disease (n, %) Rheumatic
44 (95.65 %)
Congenital
2 (4.35 %)
Position of the mechanical valve (n, %) Mitral
35 (76.09 %)
Aortic
2 (4.35 %)
Mitral ? Aortic
9 (19.57 %)
pregnancy and regularly followed up by cardiologists. The usage of warfarin was all less than 5 mg/day and daily dose was 1.25–4.375 mg, the INR was controlled between 1.6 and 2.2. No maternal thromboembolic was observed during pregnancy. Sixteen patients (34.78 %) suffered from general bleeding, in which subcutaneous ecchymosis occurred in nine patients (19.57 %), epistaxis occurred in five patients (10.87 %), and gingival bleeding occurred in two patents (4.35 %). Heart failure occurred during CS in one patent (2.17 %), but recovered after emergency treatment. Other cardiac complications occurred in 15 patients (32.61 %), in which 7 patients (15.22 %) suffered from atrial fibrillation and 8 patients (17.39 %) suffered from ventricular premature beat. 46 full-term babies were born and the mean birth weight was 2978.65 g (range 2180–4000). In these 46 neonates, there were two low birth weight infants, their weight were 2180 g and 2290 g, respectively, and were transferred to department of neonatology after birth. Apgar scores of 46 neonates at birth were more than 7 points, with no neonatal or fetal death, no stillbirth, no congenital abnormalities or neurological dysfunctions. No warfarin embryopathy, characterized by nasal hypoplasia, stippled epiphyses, or both, was observed. Thirty-eight patients received selective CS, in which 17 patients received single oral anticoagulation treatment and 21 patients received ‘‘bridging’’ anticoagulation treatment during postoperative period. There was one maternal thromboembolic event in the oral therapy group and led to death. The patient was a 21-year-old woman with a history of rheumatic heart disease and received mitral valve replacement at the age of 12. She was treated with warfarin 3.75 mg/day. Her INR level ranged from 1.6 to 2.2 throughout pregnancy and received a selective CS at the 38?2 weeks of gestation. Echocardiography before the operation demonstrated no valve thrombosis but pulmonary artery hypertension. The operation went smoothly, and the warfarin was reverted 8 h after the operation. On the third day after the operation, the patient felt chest discomfort, and in the meanwhile, her INR level was 0.89. Thereafter, she acutely became short of breath, cardiac arrest and died soon. The autopsy demonstrated mitral mechanical valve thrombosis of approximately 16 mm 9 14 mm. The time for reinitiating anticoagulation therapy after CS was 7.65 h (range 4–11) in single oral warfarin group and 7.48 h (range 4–11) in ‘‘bridging’’ treatment group, where no statistically significant difference was found between two groups (p = 0.961). There was no significant difference in the amount of uterus bleeding during 3 days after surgery between two groups. No late postpartum hemorrhage, periphery thrombosis, general hemorrhage or any other over-anticoagulation complication were observed
123
Arch Gynecol Obstet Table 2 The comparisons between single oral warfarin group and ‘‘Bridging’’ treatment group Group
n
Time for Therapy to CS (h)
Uterine Bleeding (ml) 24 h
24–48 h
48–72 h
Valve Thrombosis
Periphery Thrombosis
Bridging
21
7.48
241.67
147.24
83.10
0
0
Oral
17
7.65
235.00
140.59
79.41
1
0
p
0.787
0.661
in two groups. The comparisons between two groups are shown in Table 2. Due to premature rupture of membranes and patients refused to choose natural childbirth, eight patients received emergency cesarean Section. 10 mg vitamin k1 was given intravenously, and operation was underwent until INR reached normal. ‘‘Bridging’’ treatment was administered 7.50 h (range 4–10) after delivery and the INR level was examined daily. Serial changes in INR levels during postpartum period were compared between vitamin K1 treatment group, single oral warfarin group and ‘‘bridging’’ treatment group. There was no significant difference in time to reinitiated anticoagulation therapy between these three groups. The INR levels in vitamin K1 treatment group increased more slowly than other two groups, which need 6.38 ± 1.06 days to achieve the target range. The ‘‘bridging’’ treatment group and oral anticoagulation group needed 6.42 ± 0.81 days and 4.71 ± 0.92 days, respectively. Significant difference is shown between ‘‘vitamin K1 treatment’’ group and the other two groups (p \ 0.05), with no significant difference between the other two groups (p = 0.767). The comparison of serial changes in INR levels between three groups is shown in Fig. 1.
Discussion Pregnancy induces a series of haemostatic changes, with an increase in concentration of coagulation factors, fibrinogen, and platelet adhesiveness, as well as diminished fibrinolysis, which lead to hypercoagulability and an increased risk of thromboembolic events. In addition, obstruction to venous return by the enlarging uterus causes stasis and a further rise in risk of thromboembolism. Therefore, anticoagulation therapy is important for pregnant women who had received mechanical heart valve replacement. To avoid intraoperative and postoperative hemorrhage, the coagulation function of the patients should be maintained normal. In addition, the high maternal complication rate including valve thrombosis and resultant death should not be ignored either. Therefore, an effective anticoagulation therapy, after the bleeding tendency, is particularly important. Since the mid 1980s, heart surgeon has recognized that excessive
123
0.699
0.561
Fig. 1 The comparison of serial changes in INR between single oral warfarin group, ‘‘bridging’’ treatment group and vitamin K1 treatment group
anticoagulation strength is the important reason of postoperative bleeding and reduced the anticoagulation strength. The value of PTR has dropped to 1.5–2.0 from 3.0 to 4.0 and INR was dropped to 2.0–3.0 from 4.0 to 6.0. In this strength, the rate of bleeding was 1.4–2.4 % and the embolism was 2.0–3.8 % in western countries [10]. But in the same standard of anticoagulant strength, Chinese patients showed higher bleeding rate (0.68–10.4 %) and lower embolism rate (0.3–1.48 %) [11]. This shows that because of the difference between the races, it is necessary to use low intensity of anticoagulant in Chinese patients. And some studies also showed low-intensity anticoagulation was safe and effective for Chinese patients [8, 12]. So, the patients in our study received oral low-dose warfarin anticoagulation throughout pregnancy, with INR controlled between 1.6 and 2.2, and no complications happened, such as embolism and severe bleeding. It shows that the oral anticoagulant treatment of warfarin for pregnant women is safe and effective.
Arch Gynecol Obstet
In our study, we observed that even though the warfarin was reinitiated as soon as possible, the INR level still needed 4.71 ± 0.92 days to reach the target range. Patients who discontinued warfarin preoperatively were in the ‘unprotecting’ state, which were subjected to high risk of mechanical valve thrombosis until their postoperative INR reached the target range. In the single oral warfarin group, there was one patient suffered from mitral valve thrombosis and resulted in death. Therefore, single oral warfarin therapy might be insufficient for anticoagulation during perioperative period. LMWH works fast, and it takes only 3 h after subcutaneous injection that the peak concentration in serum is reached with almost 100 % bioavailability. Moreover, it is easier to achieve stable anticoagulation and evaluation. Moreover, LMWH is reported to have lower incidence of osteoporosis and thrombocytopenia compared with unfractionated heparin [13]. The ‘‘bridging’’ anticoagulation treatment reinitiates warfarin, and administers subcutaneous LMWH twice daily to maintain anti-Xa levels of 1–1.2 IU/ml until the target INR is reached. In our study, one out of 17 patients without postoperative ‘‘bridging’’ therapy died due to valve thrombosis and instead of increasing the morbidity of postpartum hemorrhage, late postpartum hemorrhage and general hemorrhage, the ‘‘bridging’’ treatment effectively prevented the occurrence of mechanical valve thrombosis, periphery thrombosis and stoke. Therefore, ‘‘bridging’’ treatment might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement after selective CS. Since LMWH does not influence APTT, not only INR but also anti-Xa levels should be checked daily during postoperative period. Because the anticoagulant effect of warfarin can be antagonized by vitamin K1, it is a common way to use high dose of vitamin K1 to correct coagulant disturbance when postoperative patients who received valve replacement are in over-anticoagulant state or before emergency operation. However, high doses of vitamin K1 might induce ‘‘warfarin resistance’’ [14], which could influence anticoagulant effect of warfarin even during postoperative period. In our study, it is obvious that the postoperative increase of INR in the vitamin K1 treatment group was significant more slow than other two groups. Hence, for the selective CS, the warfarin therapy should be discontinued 2 days before surgery instead of vitamin K1 application via intravenous infusion. For the emergency CS, due to the use of vitamin K1 and ‘‘warfarin resistance’’, the ‘‘bridging’’ treatment might be more appropriate and effective than single oral warfarin therapy during postoperative period until the target INR is reached. Warfarin provides the most effective control of anticoagulation with mechanical heart valves. In addition, it is considered to be the best method to prevent maternal
thrombotic events [15]. However, the best prophylaxis becomes questionable during pregnancy, for being unsafe to the fetus. Warfarin could cross the placenta and affect the fetus, which lead to ‘‘warfarin embryopathy’’, miscarriage and stillbirth [16]. Since 1991, Cotrufo et al. supported that those complications could be dose related, and suggested that warfarin would be safe as long as its use during pregnancy was limited to 5 mg/day [17]. However, the safety of low-dose oral warfarin therapy during first trimester is still under controversy [18]. In our study warfarin was continuously used throughout pregnancy and no warfarin embryopathy or over-anticoagulation in the fetus was observed, but since the objects and including criteria, this result might not reflect the real impact of warfarin on the fetus. However, there are several limitations to our study that must be considered. First, this is a retrospective, singlecenter study with a limited ability to reliably determine heart function and complications during pregnancy by patients’ records. Second, due to the limitations of knowledge and conditions, we did not use prothrombin complex concentrates but vitamin K1 to normalize the INR values prior to emergency cesarean section, which might increase the risk of patients. Third, we did not follow-up the neonatal outcomes, such as nervous system development, mental development, so this study could not reflect the impact of long-term development of fetus. Forth, the overall patient population is certainly limited, and the subgroup undergoing ‘‘vitamin K1 treatment’’ is even smaller. These limitations might be solved by increasing number of cases and participation of more research centers in future, and our study results could be confirmed by the larger sample study. And the use of prothrombin complex concentrates prior to emergency cesarean section could improve the safety of pregnant women with mechanical heart valve replacement. The discontinuation of warfarin during perioperative period increases the risk of mechanical valve thrombosis and death. The use of vitamin K1 preoperatively might result in warfarin resistance and discontinue warfarin therapy before selective CS might be more appropriate than application of vitamin K1. The ‘‘bridging’’ anticoagulation treatment which combines oral warfarin and subcutaneous LMWH might be more effective and safer than single oral warfarin therapy for patients with mechanical heart valve replacement during postoperative period, no matter selective or emergency CS. The safety of low-dose oral warfarin therapy throughout pregnancy is still under controversy, the role of us is to continue the discussion of the pros and cons of available regimens and, ultimately, help the patient and her family to decide which drug to take. We consider that the most important is the mother’s life, second is the baby’s life or avoiding embryopathy.
123
Arch Gynecol Obstet Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Limet R, Grondin CM (1977) Cardiac valve prostheses, anticoagulation, and pregnancy. Ann Thorac Surg 23(4):337–341 2. Salazar E, Izaguirre R, Verdejo J, Mutchinick O (1996) Failure of adjusted doses of subcutaneous heparin to prevent thromboembolic phenomena in pregnant patients with mechanical cardiac valve prostheses. J Am Coll Cardiol 27(7):1698–1703 3. Pieper PG, Balci A, Van Dijk AP (2008) Pregnancy in women with prosthetic heart valves. Neth Heart J 16(12):406–411 4. Chan WS, Anand S, Ginsberg JS (2000) Anticoagulation of pregnant women with mechanical heart valves: a systematic review of the literature. Arch Intern Med 160(2):191–196 5. Schaefer C, Hannemann D, Meister R, Elefant E, Paulus W, Vial T, Reuvers M, Robert-Gnansia E, Arnon J, De Santis M, Clementi M, Rodriguez-Pinilla E, Dolivo A, Merlob P (2006) Vitamin K antagonists and pregnancy outcome. A multi-centre prospective study. Thromb Haemost 95(6):949–957 6. Vitale N, De Feo M, De Santo LS, Pollice A, Tedesco N, Cotrufo M (1999) Dose-dependent fetal complications of warfarin in pregnant women with mechanical heart valves. J Am Coll Cardiol 33(6):1637–1641 7. van Driel D, Wesseling J, Sauer PJ, Touwen BC, van der Veer E, Heymans HS (2002) Teratogen update: fetal effects after in utero exposure to coumarins overview of cases, follow-up findings, and pathogenesis. Teratology 66(3):127–140 8. Bian C, Wei Q, Liu X (2012) Influence of heart-valve replacement of warfarin anticoagulant therapy on perinatal outcomes. Arch Gynecol Obstet 285(2):347–351 9. De Santo LS, Romano G, Della Corte A, D’Oria V, Nappi G, Giordano S, Cotrufo M, De Feo M (2012) Mechanical aortic
123
10.
11.
12.
13.
14. 15.
16.
17.
18.
valve replacement in young women planning on pregnancy: maternal and fetal outcomes under low oral anticoagulation, a pilot observational study on a comprehensive pre-operative counseling protocol. J Am Coll Cardiol 59(12):1110–1115 Cannegieter SC, Torn M, Rosendaal FR (1999) Oral anticoagulant treatment in patients with mechanical heart valves: how to reduce the risk of thromboembolic and bleeding complications. J Intern Med 245(4):369–374 Dong L, Shi YK (1996) Anticoagulation therapy of patients with mechanical heart valve prostheses: follow-up survey of 126 cases. Chin J Thorac Cardiovasc Surg 12:149–151 Dong L, Shi YK, Tian ZP, Ma JY, Wang X, Yi J (2003) Low intensity anticoagulation therapy after mechanical heart valve replacement. Zhonghua wai ke za zhi [Chin J Surg] 41(4):250–252 Montalescot G, Polle V, Collet JP, Leprince P, Bellanger A, Gandjbakhch I, Thomas D (2000) Low molecular weight heparin after mechanical heart valve replacement. Circulation 101(10):1083–1086 Patriquin C, Crowther M (2011) Treatment of warfarin-associated coagulopathy with vitamin K. Expert Rev Hematol 4(6):657–665 Higa T, Okura H, Tanemoto K, Yoshida K (2014) Prosthetic valve thrombosis caused by heparin-induced thrombocytopenia thrombosis during pregnancy. Circ J 78(4):1004–1005 Cade JF, Hirsh J, Martin M (1969) Placental barrier to coagulation factors: its relevance to the coagulation defect at birth and to haemorrhage in the newborn. Br Med J 2(5652):281–283 Cotrufo M, de Luca TS, Calabro R, Mastrogiovanni G, Lama D (1991) Coumarin anticoagulation during pregnancy in patients with mechanical valve prostheses. Eur J Cardiothorac Surg 5(6):300–304 Hassouna A, Allam H (2014) Limited dose warfarin throughout pregnancy in patients with mechanical heart valve prosthesis: a meta-analysis. Interact Cardiovasc Thorac Surg 18(6):797–806
