Adverse pregnancy outcomes (APOs) are often conceptualised as a window into future cardiovascular disease (CVD) risk. The available evidence for commonly occurring APOs such as hypertensive disorders of pregnancy, gestational diabetes and preterm birth suggest a two-fold higher risk of CVD following these APOs in the birthing adult.^{1, 2} In addition, APOs are associated with a higher risk of development of CVD among offspring.^{3, 4} However, longitudinal data on CVD outcomes following less commonly occurring APOs, such as placental abruption, are limited. Clarifying the unique pathways that link commonly and less commonly occurring APOs with CVD will guide tailored efforts to prevent subsequent pregnancy complications and CVD in birthing people and their offspring. Therefore, the development of large epidemiologic datasets that span several decades of the life course and across generations is needed to capture both pregnancy-related and CVD-related events. In addition, studies promoting our understanding of intergenerational transmission require access to long-term data, not only for the birthing person but for their offspring.

Identifying data sources that fulfill these criteria is challenging, particularly in the United States. In the absence of universal health care, birthing people often switch healthcare providers and/or insurers several times throughout their life course. Electronic health records and insurance claims data are often riddled with missing clinical information, limited follow-up and do not collect data on important confounders and risk factors (e.g. diet quality and physical activity levels). Longitudinal data from observational cohort studies such as the Coronary Artery Risk Development in Young Adults Study, the Women's Health Initiative and the Study of Women's Health Across the Nation offer insights into associations between APOs and CVD, however, are often limited by small sample sizes for APOs and self-reported pregnancy outcomes. Countries with universal healthcare systems and centralised healthcare databases have easier access to large populations with longitudinal data ideal for studying APOs and CVD. In fact, most of the data linking rare APOs (such as placental abruption) with a lifetime risk of CVD have been obtained from healthcare databases outside of the United States.¹

In this issue of Paediatric and Perinatal Epidemiology, Ananth and colleagues⁵ describe the methods and baseline characteristics of the Placental Abruption and Cardiovascular Event Risk (PACER) study, one of the first attempts to create a large population-based dataset of birthing people and their offspring with long-term follow-up in the United States. The study authors linked data from the New Jersey Vital Records, Hospital Discharge Data Collection System and Mortality Data from 1993 to 2020 using a probabilistic record-matching algorithm. Data were successfully linked to hospitalisation records for 92.4% of live birth records and 70.7% of fetal death records, for a total sample of 3,093,241 deliveries that occurred at or after 20 weeks' gestation. The cohort included a socio-demographically diverse population with 14.3% of deliveries to those who identified as non-Hispanic Black, 22.9% to those who identified as Hispanic, 40% to those with a high school education or less and 62.5% to those utilising private insurance.

In the dataset, there were 33,058 (1.1%) cases of placental abruption consistent with national estimates. Prevalence differed by baseline characteristics and CVD-related factors: higher rates were observed among deliveries to older birthing people, multiparas, those who identified as non-Hispanic Black, or did not have prenatal care, as well as to those who smoked during pregnancy, had pregestational diabetes or had a hypertensive disorder of pregnancy. As expected, higher rates of perinatal complications were noted among deliveries complicated by placental abruption including fetal death, fetal distress and preterm birth.

The PACER cohort represents one of the first attempts to include the birthing person and offspring in a single longitudinal dataset. We applaud the authors on this enormous undertaking, linking data from multiple sources with discrepant coding systems and personal identifiers. As acknowledged by the authors, the cohort does have several limitations. First, approximately 7.5% of all live births and fetal deaths were excluded from the final cohort due to missing data or unsuccessful data linkage. To account for possible selection bias, the authors reweighted the sample prior to analyses, however, the potential for unmeasured bias remains. Second, the PACER dataset is limited to one US state. Therefore, out-of-state CVD events will not be captured, leading to potential underestimation of incident CVD. This may be particularly concerning given rising rates of state-to-state migration within the United States; only about half of US citizens currently live in their birth state.⁶ In addition, data will only be representative of one US state. Maternal and infant outcomes vary substantially between US regions and states, reflecting underlying differences in social, structural and environmental factors influencing disease. Additional linkage of vital data, hospital records and mortality files both within and between US states will be necessary to accurately describe rates of incident CVD, and their association with APOs in the United States.

Despite these limitations, the PACER cohort has the potential to make valuable contributions to our understanding of associations between placental abruption and long-term CVD. The authors emphasise the cohort's potential to (i) clarify these associations among high-risk individuals including those with recurrent abruption and non-singleton births; (ii) identify the potential mediating effect through preterm delivery and perinatal complications and (iii) describe the moderating effect of social determinants of health. These are important steps towards unravelling the pathways that link abruption with CVD. Several questions remain, however, specifically regarding the role of other cardiovascular health factors, such as maternal blood pressure elevations, an important risk factor for both abruption and CVD.

First, what is the nature of the relationship between hypertensive disorders of pregnancy (HDP) and placental abruption in predicting later CVD risk? Are HDPs and abruption on a causal pathway to CVD whereby abruption mediates part of the association between HDPs and CVD? Or is there effect modification such that associations between abruption and CVD differ based on HDP status? Furthermore, do these relationships vary based on HDP subtypes (e.g., chronic hypertension, gestational hypertension, preeclampsia)? Given large sample sizes for each of these outcomes, the PACER cohort has the potential to shed light on each of these questions. This epidemiologic work can, in turn, guide preclinical research to clarify the pathophysiologic mechanisms by which HDPs and abruption are associated with each other and with CVD such as maternal vascular malperfusion, microvascular dysfunction, inflammation and endovascular oxidative stress.

Second, does placental abruption influence long-term blood pressure trajectories after delivery independent of blood pressure during pregnancy? Placental abruption and other APOs related to maternal vascular malperfusion are thought to trigger oxidative stress and the release of endothelial particles leading to vascular injury. Previous work has demonstrated that people with evidence of maternal vascular malperfusion have higher blood pressure in both the short- and long-term after delivery.⁷ However, whether or not the physiologic changes during pregnancy contribute to long-term vascular damage and incident hypertension, independent of underlying risk factors for vascular damage (e.g., chronic hypertension, prehypertension and microvascular dysfunction) remains unclear.

Third, could tighter blood pressure control before and during pregnancy reduce the risk of both abruption and CVD? Outside of pregnancy, robust evidence exists demonstrating a higher risk of CVD even among people with elevated (systolic and diastolic blood pressures 120–129 and <80 mmHg, respectively) and stage 1 hypertension (SBP 130–139 mmHg or DBP 80–89 mmHg); clinical guidelines recommend lowering blood pressure to <130/80 mmHg for most people with hypertension and CVD risk factors.⁸ During pregnancy, blood pressure goals are higher due to concerns regarding impaired uteroplacental blood flow. Only recently were treatment goals for chronic hypertension reduced from 160/110 to 140/90 mmHg in light of results from the Chronic Hypertension and Pregnancy trial demonstrating improvement in the composite outcome of preeclampsia with severe features, clinician-initiated indicated preterm, placental abruption or fetal or neonatal death associated with tighter blood pressure control.^{9, 10} Although individual associations between tighter blood pressure control and placental abruption were not significant, the total number of abruption events was small. Additional work is needed to investigate the effects of tighter blood pressure control of both chronic and new-onset HDPs on the risk of abruption and long-term risk of CVD.

Regardless of the pathways that link placental abruption and CVD, the promotion of cardiovascular health remains the cornerstone of CVD prevention for all birthing people. Pregnancy can serve as an ideal opportunity to screen for, counsel and promote healthy lifestyle behaviours given high levels of engagement with the healthcare system and the potential to impact, not only the birthing person but their offspring. Further unravelling the connections between APOs and CVD will inform tailored interventions to ultimately improve maternal health and prevent CVD in the birthing person and offspring.

不良妊娠结局 (APO) 通常被视为了解未来心血管疾病 (CVD) 风险的窗口。常见 APO（例如妊娠期高血压疾病、妊娠期糖尿病和早产）的现有证据表明，患有这些 APO 的成年人患 CVD 的风险增加两倍。^{1, 2}此外，APO 与后代患 CVD 的较高风险相关。^{3, 4}然而，关于不太常见的 APO（例如胎盘早剥）后 CVD 结果的纵向数据有限。阐明常见和不常见的 APO 与 CVD 之间的独特途径将指导针对性的努力，以预防分娩者及其后代随后的妊娠并发症和 CVD。因此，需要开发跨越生命历程数十年、跨代的大型流行病学数据集，以捕获与妊娠相关和心血管疾病相关的事件。此外，促进我们对代际传播的理解的研究需要获得长期数据，不仅针对出生者，而且针对他们的后代。

识别满足这些标准的数据源具有挑战性，特别是在美国。在缺乏全民医疗保健的情况下，分娩者在一生中往往会多次更换医疗保健提供者和/或保险公司。电子健康记录和保险索赔数据往往充满了缺失的临床信息、有限的后续行动，并且不收集有关重要混杂因素和风险因素（例如饮食质量和身体活动水平）的数据。来自观察性队列研究（例如年轻人冠状动脉风险发展研究、妇女健康倡议和全国妇女健康研究）的纵向数据提供了对 APO 与 CVD 之间关联的深入了解，然而，这些数据往往受到样本量较小的限制。 APO 和自我报告的妊娠结局。拥有全民医疗保健系统和集中医疗保健数据库的国家可以更轻松地获取大量人口，并获得适合研究 APO 和 CVD 的纵向数据。事实上，大多数将罕见 APO（例如胎盘早剥）与终生 CVD 风险联系起来的数据都是从美国以外的医疗保健数据库中获得的。¹

在本期《儿科和围产期流行病学》中，Ananth 及其同事⁵描述了胎盘早剥和心血管事件风险 (PACER) 研究的方法和基线特征，该研究是创建基于人口的大型分娩者及其出生数据集的首次尝试之一。后代在美国进行长期随访。研究作者使用概率记录匹配算法将 1993 年至 2020 年新泽西州生命记录、医院出院数据收集系统和死亡率数据的数据关联起来。数据成功与 92.4% 的活产记录和 70.7% 的胎儿死亡记录的住院记录关联起来，总共有 3,093,241 例妊娠 20 周或之后分娩的样本。该队列包括社会人口结构多样化的人群，其中 14.3% 的分娩对象为非西班牙裔黑人，22.9% 的分娩对象为西班牙裔，40% 的分娩对象为高中以下学历，62.5% 的分娩对象为非西班牙裔黑人。私人保险。

在数据集中，有 33,058 例 (1.1%) 胎盘早剥病例与国家估计值一致。患病率因基线特征和 CVD 相关因素而异：观察到高龄产妇、经产妇、非西班牙裔黑人或没有接受过产前护理的人以及怀孕期间吸烟的人的分娩率较高，患有妊娠前糖尿病或妊娠期高血压疾病。正如预期的那样，并发胎盘早剥的分娩中围产期并发症的发生率较高，包括胎儿死亡、胎儿窘迫和早产。

PACER 队列代表了将出生者和后代纳入单个纵向数据集中的首次尝试之一。我们对作者这项艰巨的事业表示赞赏，将多个来源的数据与不同的编码系统和个人标识符联系起来。正如作者所承认的，该队列确实存在一些局限性。首先，由于数据缺失或数据链接不成功，大约 7.5% 的活产和胎儿死亡被排除在最终队列之外。为了考虑可能的选择偏差，作者在分析之前重新加权了样本，但是，未测量偏差的可能性仍然存在。其次，PACER 数据集仅限于美国一个州。因此，州外 CVD 事件将不会被捕获，从而导致对事件 CVD 的潜在低估。鉴于美国境内州际移民率不断上升，这一点可能尤其令人担忧；目前只有大约一半的美国公民生活在他们的出生州。⁶此外，数据仅代表美国一个州。美国各地区和州之间的孕产妇和婴儿结局差异很大，反映出影响疾病的社会、结构和环境因素的根本差异。为了准确描述 CVD 事件发生率及其与美国 APO 的关联，需要将美国各州内部和之间的重要数据、医院记录和死亡率档案建立额外的联系。

尽管存在这些局限性，PACER 队列仍有可能为我们了解胎盘早剥与长期 CVD 之间的关联做出宝贵贡献。作者强调该队列具有以下潜力：(i) 阐明高风险个体之间的这些关联，包括反复性胎早剥和非单胎出生的个体； (ii) 确定早产和围产期并发症的潜在中介作用，以及 (iii) 描述健康问题社会决定因素的调节作用。这些是阐明胎盘早剥与 CVD 关联途径的重要步骤。然而，仍然存在一些问题，特别是关于其他心血管健康因素的作用，例如母亲血压升高，这是胎早剥和心血管疾病的重要危险因素。

首先，妊娠期高血压疾病 (HDP) 和胎盘早剥在预测后期 CVD 风险方面的关系本质是什么？ HDP 和胎盘早剥是否是 CVD 的因果途径，而胎盘早剥是 HDP 与 CVD 之间部分关联的中介者？或者是否存在效应修改，使得胎盘早剥和 CVD 之间的关联因 HDP 状态而异？此外，这些关系是否因 HDP 亚型（例如慢性高血压、妊娠期高血压、先兆子痫）而异？鉴于每个结果都有大量样本，PACER 队列有可能阐明每个问题。这项流行病学工作反过来可以指导临床前研究，以阐明 HDP 和胎盘早剥相互关联以及与 CVD（如母体血管灌注不良、微血管功能障碍、炎症和血管内氧化应激）相关的病理生理机制。

其次，胎盘早剥是否会影响产后的长期血压轨迹，而与怀孕期间的血压无关？胎盘早剥和其他与母体血管灌注不良相关的 APO 被认为会引发氧化应激和内皮颗粒的释放，从而导致血管损伤。先前的研究表明，有母亲血管灌注不良证据的人在产后短期和长期血压均较高。⁷然而，妊娠期间的生理变化是否会导致长期血管损伤和高血压事件，而与血管损伤的潜在危险因素（例如慢性高血压、高血压前期和微血管功能障碍）无关，仍不清楚。

第三，怀孕前和怀孕期间更严格的血压控制是否可以降低胎盘早剥和心血管疾病的风险？在怀孕之外，存在强有力的证据表明，即使在高血压（收缩压和舒张压分别为 120-129 和 <80 mmHg）和 1 期高血压（收缩压 130-139 毫米汞柱或舒张压 80-89 毫米汞柱）的人群中，心血管疾病的风险也更高。 ）；临床指南建议大多数患有高血压和 CVD 危险因素的人将血压降低至 <130/80 mmHg。⁸怀孕期间，由于担心子宫胎盘血流受损，血压目标较高。直到最近，根据慢性高血压与妊娠试验的结果，慢性高血压的治疗目标从 160/110 降低至 140/90 mmHg，该试验表明，具有严重特征的先兆子痫、临床医生发起的早产、胎盘早剥或妊娠的综合结果有所改善。胎儿或新生儿死亡与更严格的血压控制有关。^{9, 10}虽然严格血压控制与胎盘早剥之间的个体关联并不显着，但胎盘早剥事件的总数很少。还需要开展更多工作来研究严格控制慢性和新发 HDP 的血压对胎盘早剥风险和 CVD 长期风险的影响。

无论胎盘早剥和 CVD 之间的联系途径如何，促进心血管健康仍然是所有分娩人群预防 CVD 的基石。鉴于与医疗保健系统的高度接触以及不仅影响分娩者而且影响其后代的潜力，怀孕可以作为筛查、咨询和促进健康生活方式行为的理想机会。进一步阐明 APO 和 CVD 之间的联系将为量身定制的干预措施提供信息，以最终改善孕产妇健康并预防分娩者和后代的 CVD。