2025 年 89 巻 1 号 p. 77-82
Background: With advances in treatment, the prognosis for pregnancies complicated by pulmonary arterial hypertension (PAH) has been improving. However, PAH-related maternal mortality remains high compared with that due to other cardiovascular diseases. The specifics of PAH-related maternal deaths under advanced medical standards are not well understood.
Methods and Results: We used the maternal death registration system established by the Japan Association of Obstetricians and Gynecologists and reviewed 6 PAH-related maternal deaths from 2010 to 2022. All women were initially diagnosed with PAH during pregnancy or immediately after childbirth. The diagnosis of PAH tended to be delayed because symptoms were not reported to healthcare providers and/or a different disease was diagnosed. Cardiogenic shock occurred antepartum in 1 woman and during delivery or within 7 days after delivery in the other 5 women. Four women were resuscitated and started on extracorporeal membrane oxygenation. Pulmonary vasodilators were initiated in 4 women, with a median duration of 8 days from PAH diagnosis to starting medication. Right heart failure was the most common cause of maternal death, with a median duration of 16 days from PAH diagnosis to maternal death.
Conclusions: All PAH-related maternal deaths occurred in women who were diagnosed with PAH after pregnancy. Diagnosing PAH and initiating pulmonary vasodilators takes considerable time, highlighting the importance of early diagnosis and early treatment.
Pulmonary arterial hypertension (PAH) may be exacerbated in pregnant women due to changes in hemodynamics and may worsen due to female hormones. Because of the high mortality and morbidity rates for women with PAH during pregnancy and childbirth, since the first edition, the Japanese Circulation Society’s Guideline for Pregnancy and Delivery in Women with Heart Disease has consistently listed PAH as one of the “heart diseases for which it is strongly recommended to avoid pregnancy”.1
Recently, various pulmonary vasodilators have become available, and the prognosis of PAH has improved greatly. Moreover, with advances in treatment, the outcomes for pregnancies complicated by PAH are also improving. Twenty years ago, the mortality rate for women with PAH during pregnancy and childbirth was extremely high, ranging from 30% to 40%.2,3 In recent years, with the availability of these newer treatments, the maternal mortality rate in pregnancies complicated by PAH has decreased to 10–20%.4–6
In 2010, the Japan Association of Obstetricians and Gynecologists (JAOG) established a nationwide registration system for maternal deaths, and detailed data were reviewed by the Japanese Maternal Death Exploratory Committee (JMDEC), consisting of obstetricians, anesthesiologists, pathologists, forensic physicians, psychiatrists, emergency physicians, cardiologists, and a neurosurgeon. The JMDEC provides an annual report for the prevention of maternal death to improve the quality of obstetric healthcare and ultimately prevent maternal deaths. Based on the registry data, cardiovascular diseases are responsible for approximately 10% of maternal deaths in Japan, with PAH being one of the major causes.7
Recent European PAH guidelines recommend providing information and counseling about pregnancy for women with PAH of childbearing potential.8 With expanded treatment options, understanding the details of PAH-related maternal death is extremely valuable for conducting preconception counseling for women with PAH and managing pregnancies complicated by PAH. Therefore, the aim of this study was to elucidate how PAH-related maternal deaths have recently occurred in Japan using nationwide JAOG registry data.
Since 2010, the JAOG has collected information regarding maternal deaths. In each case, a report form is completed with approximately 100 questions aimed at eliciting detailed information about the clinical history and medical services supplied, along with medical records, including the anesthetic record, medical images, laboratory data, and pathological and autopsy reports.9 In this study, we used maternal mortality data registered and reviewed by the JMDEC from 2010 to 2022.
Data CollectedWe collected data on age, parity, height, weight, previous medical history, obstetric complications, onset of symptoms, the presence or absence of transfer, prenatal checkup facility, delivery facility, facility confirming death, time of delivery, mode of delivery, abnormalities during delivery, the outcome of the neonate, and time of death. We also summarized the detailed clinical course, including PAH conditions, in each case. Severe PAH is classified as New York Heart Association Class 3 or 4 if tests reveal a mean pulmonary arterial pressure ≥40 mmHg on catheterization or a tricuspid regurgitation pressure gradient ≥60 mmHg on echocardiography. Median values were calculated using Microsoft Excel®.
Ethical ConsiderationsThis study was approved by the Ethics Board of St. Marianna University School of Medicine, Kawasaki, Japan (Approval no. 6448; approval date May 28, 2024) and the JAOG (Approval no. 202407_1; approval date July 12, 2024). This investigation was conducted in accordance with the principles outlined in the Declaration of Helsinki. Informed consent was not obtained from patients or their families because this study was based on the analysis of anonymized institutional forms.
The JMDEC registered and reviewed 558 maternal deaths in Japan between 2010 and 2022. Of these, we reviewed 6 (1.1%) cases considered to be PAH-related maternal deaths. Table 1 presents the background characteristics and obstetric outcomes in each case. None of the women had been diagnosed with PAH before pregnancy. One had a medical history of mixed connective tissue disease (MCTD), but she was lost to follow-up. All cases were singleton pregnancies. Five women had prenatal health checkups in obstetrics-specific clinics or hospitals. Three had obstetric complications, such as threatened premature delivery and preeclampsia. One woman had delivered vaginally, and the other 5 had delivered by cesarean section, including 1 abortion in the second trimester and 1 perimortem cesarean section. Five of the pregnancies resulted in live births, with 2 of these 5 neonates having severe asphyxia.
Background Characteristics and Obstetric Outcomes
| Case no. |
Age (years) |
No. paritiesA |
Medical history |
BMI (kg/m2) |
Institution for prenatal checkups |
Obstetric complication |
Institution for delivery |
Timing of delivery |
Delivery mode |
Outcome of neonate |
Birth weight of neonate (g) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 20 | 0 | None | 18.6B | Obstetric clinic |
None | General hospital |
40w3d | VD | Alive | 2,700 (−1.1SD) |
| 2 | 26 | 2 | HDP in first pregnancy |
17.5B | Obstetric hospital |
Threatened premature labor |
General hospital |
35w6d | CS | Severe neonatal asphyxia (pH 6.701) |
2,402 (−0.0SD) |
| 3 | 27 | 0 | None | 23.9C | Obstetric clinic |
None | General hospital |
35w2d | CS | Alive | 2,212 (−0.1SD) |
| 4 | 37 | 1 | MCTD, SS | 24.3B | Obstetric hospital |
Thrombocytopenia | General hospital |
31w5d | Perimortem CS |
Severe neonatal asphyxia (pH 6.674) |
1,678 (−0.2SD) |
| 5 | 39 | 0 | None | 15.8B | Obstetric hospital |
None | General hospital |
16w3d | CS | Stillbirth | MD |
| 6 | 40 | 4 | None | 35B | General hospital |
GDM, preeclampsia |
General hospital |
34w4d | CS | Alive | 1,684 (−1.7SD) |
ANot including current pregnancy. BBody mass index (BMI) before pregnancy. CBMI before death. w, weeks of gestation; d, day; VD, vaginal delivery; SD, standard deviation; HDP, hypertensive disorder in pregnancy; MD, missing data; CS, cesarean section; MCTD, mixed connective tissue disease; SS, Sjögren syndrome; GDM, gestational diabetes.
Table 2 presents information regarding PAH. PAH was diagnosed antepartum, particularly in the second and third trimesters of pregnancy, in 4 women (Cases 1, 3, 5, and 6) and postpartum (within several hours after delivery) in 1 woman (Case 2). Four of these 5 women underwent right heart catheterization at diagnosis or during the clinical course, ruling out pulmonary hypertension derived from the left heart. Another case was strongly suspected to be PAH based on echocardiographic findings. All 5 women also underwent enhanced computed tomography, which ruled out pulmonary embolism. In another woman (Case 4), PAH was suspected due to a history of MCTD and sudden onset of syncope due to massive hemoptysis. Subsets of PAH were considered as PAH associated with connective tissue disease in 2 women, and idiopathic or unspecified PAH in the remaining 4 women. Five women had severe PAH at diagnosis (Cases 1–5). In Case 6, PAH worsened from mild to severe with advancing gestational age.
Clinical Courses of PAH
| Case no. |
Initial symptoms | Time of appearance of initial symptoms |
First medical facility visited after symptom appearance |
Timing of PAH diagnosis |
Findings of echocardiography/RHC |
Considered etiology of PAH |
PAH drug therapies |
Timing of starting PAH drug treatment |
Duration from PAH diagnosis to starting PAH drug treatment (days) |
Facility confirming death |
Duration from PAH diagnosis to death (days) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Dyspnea, palpitation | 39w | Obstetric clinic | Intrapartum (40w3d) | TRPG 83 mmHgA | Idiopathic/ unspecified |
None | – | – | General hospital | 6 |
| 2 | Dyspnea | Uncertain timing in antepartum period |
Obstetric clinic | Postpartum (several hours after delivery) |
Mean PAP ≥50 mmHgB | Idiopathic/ unspecified |
Macitentan, riociguat, NO, epoprostenol |
Postpartum (8 days after delivery) |
7 | General hospital | 22 |
| 3 | Dyspnea, palpitation, cough |
31–32w | Medical clinic | Antepartum (34w6d) | Estimated systolic PAP ≥80 mmHgA |
PAH-CTD (SSc) |
Macitentan, tadalafil, epoprostenol |
Postpartum (5 days after delivery) |
8 | General hospital | 42 |
| 4 | Dyspnea on exertion | 28w | Local hospital (emergency room) |
Estimation after death (31w5d) |
– | PAH-CTD (MCTD) |
None | – | – | General hospital | 0 |
| 5 | Dyspnea | 12w | Obstetric hospital | Antepartum (16w2d) | PAP 77/39 (56) mmHgB | Idiopathic/ unspecified |
Epoprostenol | Antepartum (16w2d) |
1 | General hospital | 10 |
| 6 | Dyspnea | After delivery | General hospital | Antepartum (28w3d) | Estimated systolic PAP 51→106 mmHgA |
Idiopathic/ unspecified |
Macitentan, riociguat, NO, beraprost, epoprostenol |
Postpartum (8 days after delivery) |
52 | General hospital | 66 |
AAssessed by echocardiography. BAssessed by right heart catheterization (RHC). NO, nitric oxide; PAH, pulmonary arterial hypertension; PAH-CTD, pulmonary arterial hypertension associated with connective tissue disease; PAP, pulmonary arterial pressure; SSc, systemic sclerosis; TRPG, tricuspid valve regurgitation pressure gradient. Other abbreviations as in Table 1.
The Figure shows the timing of symptom onset, admission and transfer, PAH diagnosis, initiation of medical treatment for PAH, delivery, cardiogenic shock, and maternal death. Five women experienced heart failure-like symptoms antepartum. Four of them delayed reporting their symptoms to doctors for weeks because one considered the symptoms as severe hyperemesis, and the others attributed them to late pregnancy. Another woman immediately visited a neighborhood clinic due to heart failure-like symptoms and she was initially diagnosed with respiratory tract infection. However, more than 2 weeks elapsed before she was referred to a general hospital with the suspicion of heart failure (Case 3).
Clinical courses of pulmonary arterial hypertension (PAH)-related maternal deaths. The figure shows the timing of symptom onset, admission and transfer to hospitals, PAH diagnosis, initiation of PAH drug therapy, delivery, cardiogenic shock, and death in each case. The red lines indicate days supported by extracorporeal membrane oxygenation (ECMO). CPR, cardiopulmonary resuscitation; CS, cesarean section; ECG, electrocardiogram; PDD, post-delivery day; pmCS, perimortem cesarean section; VD, vaginal delivery; w, weeks of gestation; d, day.
Five women required referral to a tertiary hospital (e.g., a perinatal medical center or experienced PAH centers) once or twice during their severe clinical course. Cardiogenic shock occurred antepartum in 1 woman with massive hemoptysis, and during delivery or within 7 days after delivery in the other 5 women. Four of the women were resuscitated and started on extracorporeal membrane oxygenation (ECMO). Pulmonary vasodilators were initiated in 4 women: on the same day of cardiogenic shock in 1 woman and after resuscitation with ECMO in the other 3 women. Three of these 4 women started treatment with pulmonary vasodilators at centers specializing in pulmonary hypertension. The median duration from the diagnosis of PAH to initiation of drug treatment for PAH (pulmonary vasodilators) was 8 days (range 1–52 days). All women received appropriate cardiopulmonary resuscitation and intensive care, but 3 died from right heart failure, 1 from PAH crisis, 1 from hypoxic encephalopathy, and 1 from airway obstruction due to massive hemoptysis. The median duration from PAH diagnosis to death was 16 days (range 0–66 days). An autopsy was performed in 1 case, revealing right ventricular hypertrophy and changes in the pulmonary artery associated with pulmonary hypertension. Moreover, multiple microthromboemboli and hemorrhagic infarction in peripheral vessels, which were more prominent than medial hypertrophy of the pulmonary arteries, were observed. It was considered that pulmonary hypertension progressed relatively rapidly, leading to peripheral pulmonary artery microthromboemboli and hemorrhagic infarctions.
A detailed examination was conducted to elucidate how PAH-related maternal deaths occur in advanced medical systems. During the 13-year nationwide registration period, 6 PAH-related maternal deaths occurred, making it the third most common cause of maternal death due to cardiovascular disease in Japan, following aortic dissection and peripartum cardiomyopathy.7 Based on the vital statistics from the Japanese Ministry of Health, Labour, and Welfare, the annual number of cases registered in the JMDEC is similar to the perinatal mortality rates.10 Therefore, our study accurately reflects the current situation of maternal deaths in Japan.
In all cases in this study, the women were initially diagnosed with PAH after pregnancy. In the Registry of Pregnancy and Cardiac Disease (ROPAC) study conducted worldwide, 6 of 7 PAH-related maternal deaths occurred in women diagnosed with PAH before pregnancy.5 In a single-center Chinese cohort, all PAH-related maternal deaths occurred in women diagnosed with PAH before pregnancy.11 In advanced medical systems, women with PAH of reproductive age can be closely monitored by specialists in PAH. These women may be well-informed about their high pregnancy risk and choose not to become pregnant or continue with pregnancy. Moreover, proactive medical treatment may prevent maternal death in a few cases of pregnancy among women previously diagnosed with PAH.12,13
Due to the severity of the condition in all cases in this study, saving lives may have been challenging. Nonetheless, 4 key issues and areas for future improvement have been identified: (1) a delay in reporting heart failure-like symptoms to healthcare providers; (2) a delay in diagnosing PAH; (3) the need for multiple transfers or consultations; and (4) a delay in starting PAH treatment.
Symptoms caused by PAH, such as shortness of breath and palpitations, are similar to those experienced during a normal pregnancy course. In the free-text section of the questionnaire and the attached summary, some women and their family members did not recognize their symptoms as abnormal until they became severe. A recent study has reported that a delayed PAH diagnosis is associated with a poor prognosis.14 Therefore, it is important to inform pregnant women, with or without comorbidities, about the normal and abnormal courses of pregnancy. Maintaining sufficient communication with healthcare providers during prenatal checkups can facilitate early consultation about symptoms.
In the clinical course of PAH for the women in this study, the first-contact doctors were obstetricians in 4 cases, a local clinician in 1 case, and an emergency physician in the remaining case. For those women who initially visited obstetricians, their chief complaints were uterine contractions, labor onset, or preeclampsia. The woman who visited a local clinician was diagnosed with an upper respiratory tract infection and treated for several weeks. Pregnant women with no history of cardiovascular disease are not expected to develop heart failure and PAH, so the condition is not considered in the differential diagnosis, leading to delayed diagnosis. The women in this study often showed abnormal vital signs, and performing cardiovascular examinations while considering cardiovascular disease is important for early diagnosis. Regardless of speciality, healthcare providers must discard the preconceived notion that pregnant women are healthy.
Five women in the study had prenatal checkups at obstetrics-specific clinics or hospitals and required referral to tertiary hospitals once or twice. In the case of another woman who had perinatal checkups at a general hospital, it took 4 weeks from the detection of PAH via screening echocardiography ordered by an obstetrician to consult a cardiologist. Establishing an easy collaborative system is desirable between institutions or departments conducting prenatal checkups and tertiary hospitals or specialists.
Pulmonary vasodilators were started in 4 women. However, more than a week, on average, elapsed from the diagnosis of PAH to the initiation of pulmonary vasodilators. Notably, 3 women had a cardiac arrest and were placed on ECMO before pulmonary vasodilators were started. Previous studies have shown that PAH-related maternal death likely occurs intrapartum and within 30 days postpartum.2 In our study, cardiopulmonary arrest occurred intrapartum and within 1 week postpartum in 5 women, consistent with previous reports. Maternal circulatory dynamics change dramatically intrapartum and for several days postpartum. Venous return increases immediately after the childbirth and placental expulsion due to rapid uterine contractions, whereas bleeding and vascular permeability increase due to delivery. There is a risk of severe obstetric bleeding, and thromboembolism can easily develop due to accelerated blood coagulation. After delivery, diuresis decreases, vascular permeability returns to normal, and fluid stored in the third space returns to the blood vessels. Therefore, in many cases, circulating plasma volume peaks several days postpartum.15 A fundamental aspect of a successful delivery for a pregnant woman with PAH is preparation beforehand.16 Some studies have reported that even when severe PAH is first diagnosed during pregnancy, safe delivery and postpartum management can be achieved by quickly initiating PAH treatment and stabilizing the patient’s condition before delivery.17 When PAH is diagnosed during pregnancy, PAH treatment should be started and optimized as soon as possible to avoid maternal death. Similarly, autopsy results for one of the cases in this study suggested that microthrombi may be involved in the progression of PAH, indicating that the use of anticoagulant therapy should also be considered.
Study LimitationsFirst, right heart catheterization was not performed in 2 of 6 cases, leaving PAH as a suspected diagnosis for these patients. Although the assessment of Case 4 is speculative, it was included in this case series for educational purposes, highlighting the importance of promptly conducting detailed cardiac evaluations, such as echocardiography, particularly in patients with a history of collagen disease. Second, the onset of initial symptoms was known only to the extent reported by healthcare providers, suggesting that symptoms may have appeared earlier with more detailed interviews. Third, an autopsy was not performed in 5 cases, leaving the detailed etiology unclear. Increasing the number of autopsies could provide valuable insights into the causes of maternal deaths. Finally, of the 558 cases registered with the JAOG, several sudden deaths with unknown causes may also be related to PAH.
In all cases of PAH-related maternal death in this study, PAH was diagnosed after the second trimester of pregnancy. In many cases there was a prolonged time from symptom onset to PAH diagnosis or the start of PAH treatment, highlighting the importance of early diagnosis and early treatment. Future research is needed to improve perinatal management for pregnant women with severe PAH and to compare outcomes between women who survive and those who die.
The authors extend their gratitude to the Maternal Mortality Review Committee and all the doctors who cooperated with the nationwide survey of maternal deaths.
The present study was supported by the Intramural Research Fund (22-B-2) for Cardiovascular Diseases of the National Cerebral and Cardiovascular Center.
The authors declare that there are no conflicts of interest.
This study was approved by the Ethics Committee of St. Marianna University School of Medicine (Kawasaki, Japan; Approval no. 6448; approval date May28, 2024) and the JAOG (Approval no. 202407_1; approval date July 12, 2024), and was conducted in accordance with the Declaration of Helsinki.
The data presented in this study are available from the corresponding author and JAOG/JMDEC upon reasonable request.
