2015 Volume 79 Issue 6 Pages 1357-1362
Background: The aim of the present work was to understand the current circumstances of maternal-death-related venous thromboembolism (MD-VTE) in Japan. We retrospectively investigated the characteristics of cases of MD-VTE, and compared past and present rates of occurrence.
Methods and Results: We examined the Japanese data for MD-VTE in 2010–2013, and compared it with that from 1991–1992. MD-VTE occurred in 17 women in 1991–1992, and in 13 women in 2010–2013. The maternal mortality ratio of MD-VTE was 0.7 per 100,000 in 1991–1992 and 0.4 per 100,000 in 2010–2013. Both the maternal mortality ratio and rate of MD-VTE in 2010–2013 deceased significantly compared with 1991–1992 (P<0.05). However, the number of cases of MD-VTE during pregnancy was 6 among 13 women (41%) in 2010–2013, but 1 in 17 women (6%) in 1991–1992, showing an increase (P<0.05). In the present study, cesarean delivery was more frequently associated with MD-VTE.
Conclusions: MD-VTE overall has decreased within the past 20 years in Japan. But, MD-VTE during pregnancy in 2010–2013 increased relative to 1991–1992. Future guidelines for prevention of VTE may need to extend beyond the perioperative period to decrease the incidence of MD-VTE. (Circ J 2015; 79: 1357–1362)
Pregnancy induces numerous physiological changes, which can increase the possibility of thrombosis (eg, increased circulating blood volume).1–3 In addition, hypercoagulable condition, reduction of fibrinolytic capacity, venous smooth-muscle relaxation, pressure from the uterus, and supine position during delivery can contribute to thrombosis. Therefore, pregnancy has been defined as a risk factor for thrombosis.4–7
Editorial p 1215
Heit et al reported an incidence of thrombosis during pregnancy and postpartum of 199.7 in 100,000 pregnancies.8 Further, they reported that deep venous thrombosis (DVT) was 5-fold more likely to occur in the postpartum period than during pregnancy (postpartum: 511.2 per 100,000 vs. pregnancy: 95.8 per 100,000), and that pulmonary embolism (PE) was 15-fold more likely to occur postpartum (postpartum: 159.7 per 10,000 vs. pregnancy: 10.6 per 10,000).8 In Japan, Kobayashi et al9,10 carried out a questionnaire survey of venous thromboembolism (VTE)-related obstetrics cases at 102 hospitals in 1991–1992 and reported a total number of 76, with a mortality rate of 13.2% (10/76 cases). The time of occurrence of VTE was during pregnancy in 17 of the 76 cases (22.4%), and postpartum in 59 of the 76 cases (77.6%).
Analysis according to delivery mode has confirmed that VTE is more likely to occur with cesarean rather than vaginal delivery.8–10 Kobayashi et al9,10 reported pregnancy-associated VTE occurrence rates of 0.003% (9/348.702 pregnancies) for vaginal delivery and 0.06% (50/87,382 pregnancies) for cesarean delivery.
As for inherited thrombophilia, the known associated factors include antithrombin deficiency, protein C deficiency, protein S deficiency, Factor V Leiden, and prothrombin gene mutation. Battinelli et al reported that among their cases of pregnancy-associated VTE, patients with a deficiency in antithrombin, protein C or protein S comprised 4.1% (95% confidence interval (Cl): 1.7–8.3%), Factor V Leiden comprised 2.1% (95% Cl: 0.7–4.9%), and prothrombin gene mutation comprised 2.3% (95% Cl: 0.8–5.3%).11 Factor V Leiden has not be reported in Japan.
Pregnancy-associated VTE has gradually emerged as a topic of further research and discussion, but there are few reports of maternal-death-related VTE (MD-VTE). The aim of the present work was to understand the current circumstances of MD-VTE by retrospectively investigating the characteristics of cases of MD-VTE, and comparing past and present rates of its occurrence in Japan.
Since 2010, information on all maternal deaths in Japan has been gathered by the Japan Association of Obstetricians and Gynecologists (JAOG). When a maternal death occurs in Japan, a detailed report is submitted to the JAOG and the individual data are analyzed by the Maternal Death Exploratory Committee (Chairman: T. Ikeda). The committee consists of 15 obstetricians, 4 anesthesiologists, 2 pathologists, 1 emergency physician and various specialists who attend review sessions each month to make annual recommendations to reduce the maternal mortality rate. The present study was performed as part of a series analyzing maternal deaths in Japan by this committee. We extracted the data concerning MD-VTE in 2010–2013 from the database.
Data on maternal background, including maternal age, body height and weight, body mass index (BMI), inherited thrombophilia (antithrombin deficiency, protein C deficiency, protein S deficiency, Factor V Leiden, and prothrombin gene mutation), medication during pregnancy, maternal complications, obstetric complications, familial history of VTE, time of occurrence, and episodes of bed rest, were collected. In cases of postpartum VTE, delivery mode and thromboprophylaxis after delivery were investigated.
For cases of MD-VTE during pregnancy (2010–2013), further details about the inciting episode and risk factors for VTE were investigated.
For analysis of prior cases of MD-VTE we used the data gathered by Nagaya et al, who analyzed all maternal deaths in Japan from 1991 to 1992.12 Points of comparison included age, BMI, time of occurrence, delivery mode, maternal and obstetric complications, time from onset to cardiac arrest, and maternal mortality ratio. The maternal mortality ratio was defined as the value of all maternal deaths divided by the sum of all live births and fetal deaths, multiplied by 100,000.
In both temporal groups, MD-VTE was defined as VTE demonstrated by contrast enhanced computed tomography, pulmonary arteriography, lung scintigraphy, ultrasound sonography, autopsy imaging, and/or autopsy, and either a number of expert obstetricians in the Maternal Death Exploratory Committee (2010–2013) or Nagaya’s group (1991–1992) judging VTE as the cause of death. Time of occurrence was defined as the time symptoms first occurred.
Univariate analysis was performed using the chi-squared test, and the Mann-Whitney U-test and paired t-test were used for statistical analysis. P<0.05 was considered significant.
The total number of maternal deaths from January, 2010 through December, 2013 was 184. The cause of maternal death was VTE in 13 women (7.0%). Maternal background data for the MD-VTE cases are shown in Table 1. The median maternal age was 33 years (27–43 years), median maternal height was 158 cm (153–162 cm), median maternal weight was 62 kg (42–80 kg) and median BMI was 24 (16–33). Inherited thrombophilia was unknown in any of the women. None were taking medication that would increase the risk of thrombosis. Diseases complicating maternity were epilepsy in 2 women, but medication for epilepsy was discontinued before pregnancy, as long as the patients remained in good condition. Two women had a familial history of thrombosis; in 1 case, the sister had experienced thrombosis and in the other, an uncle had had a juvenile brain infarction. The period during which VTE occurred was pregnancy in 6 women (46.1%), postpartum in 6 women (46.1%), and after induced abortion in 1 woman (7.6%). None of the women experienced VTE during the 2nd trimester of pregnancy (Figure 1). In all postpartum cases, VTE occurred within 4 days after delivery: delay of early ambulation after delivery in 4 cases; bed rest for a strained muscle in 1 case; bed rest and dehydration because of hyperemesis in 2 cases; long-term bed rest because of threatened premature labor (TPL) in 1 case(Table 2).
| Case no. | Age (years) | Height (cm) | Weight (kg) | BMI | Inherited thrombophilia |
Medication | Maternal complication |
Obstetric and gynecologic complications |
Familial history of VTE |
Time of occurrence | Episode of bed rest and/or dehydration |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 28 | 156 | 70 | 29 | Unknown | – | – | CPD | − | Postpartum 3rd day | + |
| 2 | 29 | 156 | 80 | 33 | Unknown | – | – | − | + | 14 weeks of pregnancy | + |
| 3 | 27 | 153 | 67 | 29 | Unknown | – | – | NRFHR | Unknown | Postpartum 2nd day | + |
| 4 | 39 | 162 | 62 | 23 | Unknown | – | – | CPD | Unknown | Postpartum 4th day | + |
| 5 | 29 | 159 | 66 | 26 | Unknown | – | – | − | Unknown | Postpartum 30 min | − |
| 6 | 43 | 160 | 62 | 24 | Unknown | – | – | − | Unknown | Post induced abortion 4th day | + |
| 7 | 39 | Unknown | Unknown | Unknown | Unknown | – | – | − | Unknown | 9 weeks of pregnancy | + |
| 8 | 33 | 153 | 53 | 23 | Unknown | – | Epilepsy | − | Unknown | 39 weeks of pregnancy | − |
| 9 | 27 | 156 | 51 | 21 | Unknown | – | Epilepsy, Cardiovascular disease |
− | + | 12 weeks of pregnancy | + |
| 10 | 34 | 158 | 60 | 24 | Unknown | – | – | – | − | Postpartum 1 st day | − |
| 11 | 35 | 161 | 62 | 23 | Unknown | – | – | Myoma | Unknown | Postpartum 4th day | + |
| 12 | 33 | Unknown | Unknown | Unknown | Unknown | – | – | Myoma | − | 37 weeks of pregnancy | − |
| 13 | 32 | 161 | 42 | 16 | Unknown | Steroid | – | TPL | Unknown | 33 weeks of pregnancy | + |
BMI, body mass index; CPD, cephalopelvic disproportion; MD, maternal-death-related; NRFHR, non-reassuring fetal heart rate; TPL, threatened premature labor; VTE, venous thromboembolism.
Time of occurrence of maternal-death-related venous thromboembolism.
| Case no. | Inciting episode | Risk factors of VTE | |||
|---|---|---|---|---|---|
| Obesity | Smoking | Late pregnancy | Other | ||
| 2 | Bed rest because of muscle strain |
+ | – | − | – |
| 7 | Bed rest and dehydration because of hyperemesis |
Unknown | Unknown | + | – |
| 8 | – | − | Unknown | − | Protein someone deficiency suspected |
| 9 | Bed rest and dehydration because of hyperemesis |
− | Unknown | − | Cardiovascular disease |
| 12 | Dehydration because of common cold |
+ | – | − | Myoma |
| 13 | Bed rest because of TPL | − | Unknown | − | Prescribed steroids for mastocytoma |
Abbreviations as in Table 1.
The inciting episode for MD-VTE and risk factors for VTE during pregnancy (2010–2013) are shown in Table 2; 5 of 6 women (83.3%) had had an episode of bed rest and/or dehydration; 5 of 6 women (83.3%) had had a risk factor for VTE. In case 8 in which there was no inciting episode for developing MD-VTE, the patient was suspected to have protein S deficiency because the value measured at 11 weeks of pregnancy was 28%. The patient in case 2 was obese; case 7 was an older woman; in case 9 there was cardiac depression after Kawasaki’s disease; the patient in case 12 was obese and had myoma (height and weight unknown, but obesity suspected from medical record); in case 13 the patient had taken steroids for mastocytoma.
The delivery mode of the postpartum MD-VTE cases was cesarean delivery in 5 of 6 women (83.3%) and vaginal delivery in 1 of 6 women (16.6%). In 3 of 6 women (50.0%), DVT after delivery was prevented by the use of compression stockings (cases 1, 3) or intermittent pneumatic compression (case 4). Heparin was not used to prevent DVT in any case (Table 3).
| Case no. | Mode of delivery | Thromboprophylaxis | Method of prophylaxis | ||
|---|---|---|---|---|---|
| Compression stockings |
Intermittent pneumatic compression |
Heparin | |||
| 1 | Cesarean delivery | − | − | − | − |
| 3 | Cesarean delivery | + | + | − | − |
| 4 | Cesarean delivery | + | + | + | − |
| 5 | Vaginal delivery | − | − | − | − |
| 10 | Cesarean delivery | + | + | − | − |
| 11 | Cesarean delivery | + | + | − | − |
Abbreviations as in Table 1.
Comparisons of maternal deaths and MD-VTE in 2010–2013 with those in 1991–1992 are shown in Figure 2 and Table 4. The sum of all live births and fetal deaths in Japan was 2,423,923 in 1991–1992 and 4,291,452 in 2010–2013. Total number of maternal deaths in Japan was 197/2,423,923 in 1991–1992 and 184/4,291,452 in 2010–2013. Total number of MD-VTE cases in Japan was 17/2,423,923 in 1991–1992 and 13/4,291,452 in 2010–2013. The maternal mortality ratio was 8.1 per 100,000 population in 1991–1992 and 4.2 per 100,000 population in 2010–2013. The maternal mortality ratio of MD-VTE was 0.7 per 100,000 population in 1991–1992 and 0.4 per 100,000 population in 2010–2013. Both the overall maternal mortality ratio and the rate of MD-VTE in 2010–2013 showed significant decreases from 1991–1992 (P<0.05). However, the rate of VTE among maternal death cases remained static: 8.6% (17 of 197 women) in 1991–1992 and 7.0% (13 of 184 women) in 2010–2013. The BMI in 2010–2013 decreased from that in 1991–1992 (P<0.05). MD-VTE during pregnancy was 6/13 (46.6%) in 2010–2013, and 1/17 (5.8%) in 1991–1992. In 2010–2013; the occurrence of MD-VTE during pregnancy increased compared with 1991–1992 while the occurrence of MD-VTE in the postpartum period decreased (P<0.05).
Comparison of overall maternal mortality rates and maternal-death-related venous thromboembolism (MD-VTE) in 1991–1992 and 2010–2013.
| 1991–1992 (n=17) | 2010–2013 (n=13) | |
|---|---|---|
| Age (years) | 30 (24–40) | 33 (27–43) |
| BMI | 30.5 (19.1–38.6) | 24.2 (16.4–33.2)* |
| Occurrence time | ||
| During pregnancy | 1 (5.8%) | 6 (46.1%)* |
| During delivery | 0 (0%) | 0 (0%) |
| Postpartum | 16 (94.1%) | 7 (53.8%)* |
| Cesarean delivery | 13/16 (81.2%) | 5/6 (83.3%) |
| Obstetric and gynecologic complications | ||
| Pregnancy-induced hypertension | 3 (17.6%) | 0 (0%) |
| Placental abruption | 2 (11.7%) | 0 (0%) |
| Threatened premature delivery | 1 (5.8%) | 1 (7.6%) |
| Myoma of the uterus | 2 (11.7%) | 2 (15.3%) |
| Median time from onset to cardiac arrest (min) | 20 (0–850) | 60 (0–11,520) |
*P<0.05. Abbreviations as in Table 1.
The present study investigated the characteristics of cases of MD-VTE in Japan and compared current statistics (2010–2013) with those of the past (1991–1992). MD-VTE occurs more frequently in the postpartum period, especially after cesarean delivery. MD-VTE during pregnancy occurred mainly in the 1st and 3rd trimesters. The maternal mortality ratio of MD-VTE decreased from 1991–1992 to 2010–2013, but the rate of MD-VTE during pregnancy increased. The decrease in postpartum cases of MD-VTE was the reason for the overall decrease in MD-VTE. MD-VTE has decreased, but the proportion of VTE cases among total maternal deaths did not differ between 1991–1992 and 2010–2013.
The period of onset of MD-VTE appeared to have 3 peaks in the present study: the 1st and 3rd trimesters, and postpartum. These results are similar to findings regarding the onset of VTE in general.11 VTE onset in the 1st trimester could be related to dehydration from hyperemesis or bed rest because of TPL. VTE in the 3rd trimester is believed to be the result of prolonged immobilization because of complications of severe preeclampsia, TPL, or the pregnacy is multiple (twin or more).11 Risk of VTE is reported to be 22-fold higher with cesarean delivery than with vaginal delivery.11 In the present study, cesarean delivery was more frequently associated with MD-VTE (cesarean, 4; vaginal, 1).
The incidence of MD-VTE in Japan has decreased within the past 20 years. The Japanese Society on Thrombosis and Hemostasis developed the Japanese Guideline for Prevention of Venous Thromboembolism in 2004, and VTE has now been universally recognized as significant complication after delivery in Japan. Second, use of anticoagulant therapy such as unfractionated heparin as a thromboprophylaxis after delivery. However, we found that MD-VTE had occurred in cases of thromboprophylaxis for DVT, so further decreases in MD-VTE may depend on improvements in the guidelines for thromboprophylaxis in the field of obstetrics.
The rate of MD-VTE during pregnancy in 2010–2013 increased relative to that in 1991–1992 and in almost all cases the women had an inciting episode for the development of MD-VTE. They also had the risk factors for VTE. Therefore, particularly for pregnant women with potential inciting episodes for the development of MD-VTE in addition to risk factors for VTE, and a thrombotic predisposition, we should carefully manage their pregnancy to prevent significant VTE. The present results suggest that despite advancements in reproductive medicine, the increasing number of late pregnancies is a factor.13 Therefore, the number of pregnant women with risk factors of VTE may also increase. Future guidelines for the prevention of VTE may need to extend beyond the perioperative period to decrease the incidence of MD-VTE.
We cannot be certain of the rationale underlying the decision to perform cesarean delivery, because this study was a retrospective analysis. However, these are official statistics because almost all obstetricians in Japan will have participated in the study through their association with JAOG.
In the present study, there were a number of cases (11 cases) in which the woman’s BMI was less than 25. The mean BMI in MD-VTE cases in 1991–1992 was 30.9. Moreover, the odds ratio for VTE with a BMI ≥27 was 3.47 in the report by Kobayashi et al,9,10 whereas that for a BMI <25 was 1.89. Obesity has been reported as a risk factor for the development of VTE.14,15 In the obese, the pumping function of skeletal muscle contraction is reduced, so blood stagnation can occur. On the other hands, Heit et al reported that only obesity without other complications was not a risk factor for developing VTE.16 Generally, thin people are considered not at risk of VTE.14,15 However, for a pregnant woman, being thin reduces the amount of perivascular fat and the inferior vena cava can become vulnerable to being compressed by uterus, thus predisposing the woman to supine hypotensive syndrome.17
In the present study, inherited thrombophilia was not examined, but the possibility that it plays a role in MD-VTE is incontrovertible. It will be necessary to investigate inheritance and the primary causes of thrombosis in future cases of MD-VTE.
The ratio of MD-VTE in Japan has decreased within the past 20 years, but the rate of MD-VTE during pregnancy in 2010–2013 had increased relative to the rate in 1991–1992. These results suggest that despite advancements in reproductive medicine, the increasing numbers of late pregnancies is a causative factor. Pregnant women with risk factors of VTE may be increase in number, as will the number of cesarean deliveries. In the future, the Guideline for Prevention of VTE may need to extend beyond the perioperative period to decrease the incidence of MD-VTE.
We give special thanks to all the doctors of the facilities reporting maternal deaths. This work was supported by the Japan Association of Obstetricians and Gynecologists.
