Introduction

Acute aortic dissection (AD) is a rare catastrophic cardiovascular disease but has high morbidity and mortality rates if they succumbed prior to receiving the necessary treatment.1–3 It has been demonstrated that haemodynamic and hormonal changes of pregnancy increase the risk of progressive dilatation or dissection of the aorta, and then trigger AD, in addition to increasing the risk of venous thromboembolism, myocardial infarction and stroke.4–6 Concomitance with pregnancy further makes both the mother and fetus riskier.

As guidelines consider as a type I recommendation that patients experiencing an acute type A aortic dissection during first or second trimester of pregnancy urgent aortic surgery with fetal monitoring is recommended, and during third trimester urgent caesarean delivery immediately followed by aortic surgery is recommended.7 The conventional concept of pregnancy-centred management strategy often influences surgeons’ readiness to include fetal viabilities in surgical repair of AD. As a result of the recent vision towards improving the safety of both mother and fetus, maternal survival and fetal viability are often used as a yardstick for evaluating treatment quality, and simultaneous preservation of mother and fetus is regarded as a hallmark of high-quality treatment protocol of pregnancy-related AD.7 8

However, there are currently no global consensus or guideline for standardised management of pregnancy-related AD. As per the Standford classification, all type A ADs are treated as immediate,9 however, it remains a great therapeutic challenge when dealing with the pregnant women, especially in the first and second trimesters. This paucity of well-established therapeutic guidelines has greatly contributed to the uncertainty regarding the optimal strategy of simultaneously preserving both the pregnant woman and the fetus in case of pregnancy-related AD.10–12

The main purpose of this study was to investigate the clinical features and outcomes of women with acute AD related to pregnancy and to facilitate the optimal therapeutic strategy of pregnancy-related AD by ascertaining the currently actual perioperative and perinatal outcomes of pregnant women and fetus.

Patients and methodsStudy design

This multicentre cohort study was conducted by the Additive Anti-inflammatory Action for Aortopathy & Arteriopathy (5A) Investigators.13–15 5A study was a national registry study in which eligible patients with aortic dissection consecutively enrolled at China cardiovascular centres since 2016, aiming to monitor and improve the quality of AD treatment in mainland China. Focusing on optimising therapeutic strategy of pregnancy-related AD in our present study, we retrospectively identified consecutive adult women patients aged 18 years or older who experienced an acute AD during pregnancy or postpartum (≤12 weeks) diagnosed by aortic CT angiography from Chinese three representative large-volume cardiovascular centres (Beijing Anzhen Hospital Affiliated with Capital Medical University (from 1 January 2003 to 31 December 2021), Shanghai Zhongshan Hospital Affiliated with Fudan University (from 1 January 2013 to 31 December 2021) and Guangzhou Provincial Hospital (from 1 January 2004 to 31 December 2021)) (figure 1).

Figure 1

Patient selection flowchart.

We included patients with an acute AD diagnosis and were admitted to hospitals within 14 days of symptom onset, based on the clinical practices and existing literature. Patients were excluded if they had missing data on age, gestational week or perinatal and perioperative outcomes; had traumatic dissection, iatrogenic dissection and onset time>14 days. AD is classified according to the Stanford system.16 Type A AD is defined as the dissection involving ascending aorta while type B AD is defined by an entry tear beyond the origin of the left subclavian artery. The study was approved by the Institutional Review Board of Beijing Anzhen Hospital Affiliated with Capital Medical University, Institutional Review Board of Shanghai Zhongshan Hospital Affiliated with Fudan University and Institutional Review Board of Guangzhou Provincial Hospital. Waiver/exemption of informed consent was granted by each Institutional Review Board due to this retrospective nature. This study was conducted according to the Declaration of Helsinki and comparable Chinese ethical standards. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline.17

Data source

All patient’s information was collected by electronic medical records system or manual medical records. Baseline and demographic characteristics included age, weight, height, body mass index and body surface area. Clinical profiles and risk factors included gestational week at onset, time from onset to confirmed, time of onset of AD, birth parity and hypertension, family history of aortic disease, pre-eclampsia, eclampsia, diabetes mellitus, previous cardiac history and Marfan syndrome. Cardiovascular characteristics on admission included aortic sinus diameter with its Z value, sinus diameter, left ventricular end-diastolic dimension, left ventricular ejection fraction, aortic regurgitation degree, pericardial effusion, NYHA class, DeBakey class and location of primary entry site. Procedural characteristics included maternal–fetal management strategy, aortic therapeutic strategy, mode of delivery, timing of delivery, surgical procedures (root procedures, arch procedures, cardiopulmonary bypass (CPB) time, aortic cross-clamp time, circulatory arrest time, circulatory arrest temperature, circulatory arrest of lower body and concomitant procedures). Selective surgery is defined as an operation that needs to be completed within a period of time (1 month), and the time of the operation can be selected, but it cannot be delayed for too long. To ensure high quality of data, medical staff received training on data entry into the interactive, web-based database of Hospital Information System. A structured validation was performed periodically to check the completeness and plausibility of the data.

Our special CPB protocol included (1) using warmer temperatures (normothermia or mild hypothermia) during CPB, (2) minimising CPB times, (3) avoiding circulatory arrest before 28 gestational weeks, (4) maintaining a high flow rate (>2.4 L/m2 per minute), (5) mean arterial pressures exceeding 70 mm Hg. Since 2009, the uterus is routinely managed by insertion of a Cook balloon after caesarean delivery instead of a hysterectomy.

Outcomes

The primary outcome was a composite of maternal death and fetal death, miscarriage, stillbirth, perinatal death or neonatal death, regardless of cause, occurring within 30 days in or out of the hospital admission, after 30 days during the same hospitalisation, or subsequently corresponding perinatal stage, using International Classification of Diseases-10. The individual components of mortality were measured as secondary outcomes, including maternal death and fetal death, miscarriage, stillbirth, perinatal death or neonatal death as mentioned above. In addition, secondary outcomes included maternal mechanical ventilation time, intensive care unit stay and neonatal birth weight and Apgar score at 5 min. After discharge, the vital status of each participant was followed-up every 6 months through individual computer-based patient record, supplemented by annual active confirmation through telephone questionnaire or WeChat interview, or by regular checks against local residential and administrative records. Last follow-up was defined as last questionnaire returned, death or loss to follow-up. Follow-up continued until loss of follow-up, death or the last follow-up date in December 2023.

Statistical analyses

For descriptive statistics, frequencies with percentages were reported for categorical variables; means with SD were reported for continuous variables. If the distribution of variable was not normally distributed based on the normality test, the medians and IQR were reported. Univariate and multivariable Poisson regression were used to evaluate associations with outcomes with risk ratio (RR) with a 95% CI. Alive patients, including those who were lost to follow-up, were censored at their last follow-up date. All statistical analyses were performed using R V.3.6.1 (R Foundation, Vienna, Austria).

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

ResultsPatients’ characteristics

A total of 67 patients were included for final analysis including 42 from Beijing cohort, 11 form Guangzhou cohort and 14 form Shanghai cohort. These women experienced acute AD at age of median 31 (IQR 28–35) years (range, 18–44 years). Type A AD occurred in 46 women (68.7%) and type B AD occurred in 21 women (31.3%). Patients had onset of AD at gestational age of median 32 (IQR 25–36) weeks (range, 7 gestational weeks to 42 days postpartum). Sixty-one occurred during pregnancy (3, 17 and 41 in the first, second and third trimester, respectively) and six occurred postpartum of a median of 24 (IQR 14–33) days. For alive individuals at discharge, follow-up was complete in 98.3% (59/60) of maternal outcomes and in 98.0% (48/49) of fetal or neonatal outcomes for a median duration of 5.3 (IQR 3.0–9.0) years. There were three maternal deaths and one infant death at follow-up. Additional information is shown in table 1.

Table 1

Patient characteristics of pregnancy-related aortic dissection

Strategy and consequences

Overall, 1 was treated medically alone, 10 had endovascular procedures and 52 were treated surgically, besides four sudden deaths without any cardiovascular treatment (figure 2).

Figure 2

Practical therapeutic strategy and outcomes of pregnancy-related aortic dissection. (A) Practical therapeutic strategy and maternal outcome. (B) Practical therapeutic strategy and fetal outcome. Fetal deaths included fetal miscarriage, stillbirth, perinatal death or neonatal death, regardless of cause, occurring within 30 days in or out of the hospital admission, after 30 days during the same hospitalisation, or subsequently corresponding perinatal stage. AD, aortic dissection.

Among 40 type A AD during pregnancy, 35 received immediate aortic surgery while two received selective aortic surgery, in addition to three sudden deaths before initiating any maternal–fetal interventions. One in the first trimester terminated pregnancy first and underwent concomitant immediate aortic surgery. Of six in the second trimester, three terminated pregnancy and underwent concomitant immediate aortic surgery, two underwent immediate aortic repair and carried the pregnancy to term and one underwent immediate caesarean and subsequently selective aortic surgery. Of 33 in the third trimester, 29 received immediate caesarean and concomitant aortic surgery, 1 underwent immediate caesarean and selective aortic surgery and 3 had sudden deaths abovementioned (table 2, online supplemental figure 1). Besides, six postpartum women with type A AD all received immediate aortic surgery (online supplemental figure 3).

Table 2

Therapeutic strategies and procedures of pregnancy-related aortic dissection

All 21 type B AD occurred during pregnancy, of which 1 was treated medically alone, 10 had endovascular procedures, and 9 were treated surgically, except 1 sudden death immediately after caesarean section. Of two in the first trimester, both terminated pregnancy and subsequently underwent selective aortic surgery. Of 11 in the secondary trimester, 7 terminated pregnancy and then underwent selective aortic surgery, 2 terminated pregnancy and underwent concomitant immediate aortic surgery and 2 underwent immediate aortic surgery and carried the pregnancy to term. Of eight in the third trimester, five underwent immediate caesarean and then subsequently selective aortic surgery, one underwent immediate caesarean and concomitant aortic surgery, one received medical treatment alone and carried the pregnancy to term and one received immediate caesarean first and then suddenly died immediately after caesarean (table 2, online supplemental figure 3). Representative individual images of type A and type B aortic dissection are shown in figure 3.

Figure 3

Representative CTA findings in the pregnancy-related type A and type B aortic dissection. A1/B1: preoperative three-dimensional CTA figures of pregnancy-related type A aortic dissection group; A2-4/B2-4: preoperative, 30-day postoperatively and at 6 months of follow-up CTA figures (sagittal plane above and transverse plane below) of type A and type B aortic dissection group. CTA, computed tomography angiography.

Maternal and fetal outcomes

Overall, we observed a total of 7 maternal deaths (6 from type A aortic dissection and 1 from type B aortic dissection) and 18 fetal deaths (7 from type A AD and 11 from type B AD) (table 3). Among the maternal deaths related to type A AD, three occurred suddenly on hospital arrival, two were postoperative deaths resulting from multiple organ dysfunction following immediate caesarean section and selective aortic repair and one was due to respiratory failure following immediate caesarean and concurrent aortic surgery (online supplemental figure 1). Of the seven fetal deaths associated with type A AD, four were a result of pregnancy termination, while three occurred simultaneously with maternal sudden deaths (online supplemental figure 1). For type B AD, all 11 fetal deaths resulted from pregnancy termination, with 7 of these cases linked to signs of placental malperfusion (see online supplemental figure 2).

Table 3

Primary and secondary outcomes of pregnancy-related aortic dissection

Secondary outcomes

Patients experiencing an acute type A AD had longer mechanical ventilation time (25 (IQR 22–48) vs 2013–20 hours, p=0.037) and ICU stay (70 (41–109) vs 2012–20 hours, p<0.001) than those experiencing acute type B AD. Additional perioperative and perinatal outcomes are shown in table 3.

Risk comparisons in AD types and timing of surgery

After adjustment for significant variables (age, body mass index, gestational week, hypertension, family history of aortic disease, eclampsia, diabetes mellitus, previous cardiac history, Marfan syndrome) in a logistic regression model, the risk of type B aortic dissection compared with type A aortic dissection was significantly higher regarding composite outcome (adjusted RR: 5.03 (95% CI 1.63 to 15.58); p=0.005) and fetal outcome (adjusted RR: 6.70 (95% CI 2.00 to 22.47); p=0.002), however, was similar regarding maternal outcome (adjusted RR: 0.33 (95% CI 0.04 to 2.97); p=0.325) (figure 4A).

Figure 4

Risk of type A versus type B AD and of immediate versus elective surgery. AD, aortic dissection; RR, risk ratio.

After similar adjustment for significant variables, the risk of selective surgery compared with immediate surgery was significantly higher regarding composite outcome (adjusted RR: 12.47 (95% CI 3.26 to 47.73); p=0.0002) and fetal outcome (adjusted RR: 8.77 (95% CI 2.33 to 33.09); p=0.001), however, was similar regarding maternal outcome (adjusted RR: 6.28 (95% CI 0.52 to 75.31); p=0.147) (figure 4B).

Discussion

To the best of our current knowledge, this is the largest sample size of pregnancy-related AD in China. This study highlights the high risk of death for both pregnant women and their fetuses if they do not receive timely treatment for pregnancy-related AD. Surgical treatment with the fetus in utero should be considered before the fetus is viable, and immediate aortic surgery is recommended for type A AD at any stage of pregnancy or postpartum.

Acute AD during pregnancy and puerperium is a rare but catastrophic cardiovascular emergency,18 with devastating consequences for both parent and fetus as evidenced by significant maternal and fetal mortality, as high as 30% and 50%, respectively.19 Data from National Inpatient Sample database showed an increasing incidence of pregnancy-related AD from 2002 to 2017 in the USA,2 emphasising the need for early identification and aggressive prevention.

Haemodynamic, hematologic and metabolic changes during pregnancy impose substantial challenges to the maternal cardiovascular system, which can result in pregnancy-associated vascular remodelling, such as unphysiological dilatation and compromised integrity of the aortic wall.20–26 Consistent with the previous studies,2 the sinus of aorta was remarkably dilated in pregnant women with type A AD. For pregnant women with dilated thoracic aorta, echocardiography should be performed to track the changes of aortic root and ascending aorta in women with a predisposing aortopathy, most commonly Marfan disease.27 Selective surgery should be postponed until after delivery if possible, however, pregnancy itself should not preclude the necessity for surgery.

Recently, Martino et al conducted a meta-analysis and reported a pooled maternal and fetal mortality of 23% and 27% among patients with AD during pregnancy and puerperium, of whom 67% was type A AD.28 In contrast, maternal and fetal mortality of 14.0% and 15.2% among 46 patients with type A patients with AD in our study, including three patients who succumbed prior to receiving the necessary treatment. The main reason of discrepancy might be a result from the fact that patients with type A AD in our study had higher percentage of immediate surgery. Besides, the inconsistent difference in mortality rates between studies could suggest the differences in study group selection, publication bias, variations in medical decision-making between hospitals and countries, and whether prehospital mortality was included. Comparison with previous series,12 28 our cohorts had higher percentage of fetal demise in type B AD, the main reason of which might be due to the higher rate of pregnancy terminations for patients with type B AD. Instead of pregnancy-centred strategy, growing vision towards improving maternal survival and fetal vitality, simultaneous preservation of both mother and fetus has been considered as a hallmark of high-quality medical care in the treatment of pregnancy-related AD. However, it remains a therapeutic challenge for simultaneous maternal–fetal preservation. Significantly, timing of aortic surgery in pregnancy is a matter of current concern. However, current recommendations are based on a few, mostly retrospective studies and case reports with low evidence level, underscoring the lack of robust evidence in this field.

The first trimester and early second trimester present a dilemma for surgeons and obstetricians due to the risks including fetal loss, neonatal prematurity and maternal mortality.29 Despite the significant risk of causing fetal developmental anomalies with surgery, immediate aortic surgery was recommended for type A AD due to the high risk of death for the pregnant woman if treatment is delayed, with continuation of pregnancy. Full-course fetal monitoring and a multidisciplinary assessment of fetal viability are essential, providing valuable feedback for deciding whether to continue or terminate the pregnancy. AD during the second trimester is particularly challenging due to the imbalance between maternal haemodynamics and fetal development. During the late second trimester, management strategies mainly depend on fetal viability, which may be influenced by the capabilities of the local neonatal service. A multidisciplinary assessment is crucial to determine fetal viability. If the fetus has a high chance of survival, an immediate caesarean section followed by aortic surgery might be the best option. If not, immediate aortic surgery with the continuation of the pregnancy is an alternative, with significant emphasis on fetal protection during CPB. In the third trimester, while there are significant risks of maternal and fetal circulatory collapse, fetal development is largely complete, and fetal viability is significantly improved. Therefore, a caesarean section followed by aortic repair may offer the best chance for both fetal and maternal survival.30 On the whole, maternal fate is associated with the timing of surgery (immediate or selective) for type A AD, while fetal fate is associated with the choice of pregnancy (continuation or termination) especially for pregnant women with acute AD in the first or second trimester. However, there is no universal solution to deal with this rare but challenging catastrophe, and an optimal strategy should consider the dissection type, gestational age and fetal vitality as well as the local medical service level.

Strengths and limitations

Our study spans a period of 18 years, which allows for a more comprehensive and longitudinal analysis of patterns. Furthermore, our study includes data from three large-volume cardiovascular centres across China (Beijing, Shanghai and Guangzhou), increasing the overall representativeness and robustness when analysing clinical characteristics and outcomes. However, retrospective data collection was limited by major missingness of phenotypic features, and we failed to obtain genetic aortopathy diagnosis. The exact gestational week to decide pregnancy continuation or termination in our proposed protocol was determined based on our clinical practice and multidisciplinary consultation; however, it is unknown whether this protocol can be extrapolated to other regions of the world, and it must be interpreted in the context of Chinese pregnancy women and the current medical service level.

Conclusion

Pregnancy-related AD poses a high risk of death for both mother and fetus if not treated promptly and properly. Surgical treatment with the fetus in utero should be considered before the fetus is viable, and immediate aortic surgery should be done for type A aortic dissection, regardless of at any trimester during pregnancy or postpartum. Optimal management strategy should consider the dissection type, gestational age and the level of medical services as well as a specific longitudinal assessment of fetal vitality.