Cardiovascular mortality in the context of hypertensive disorders of pregnancy: Towards an optimisation of case identification strategies

Cardiovascular conditions have become the leading cause of maternal mortality in high-income settings.1 Hypertensive disorders of pregnancy (HDP), occurring in 10% of pregnancies worldwide, remain major contributors to pregnancy-related cardiovascular morbidity.1, 2 Thus, a better understanding of the contribution of HDP to pregnancy-related cardiovascular complications is critical to reducing the incidence of maternal morbidity and mortality going forward. In their population-based cohort study, Lee and colleagues3 explored the association between HDP subtypes (i.e. chronic hypertension, gestational hypertension, preeclampsia with mild features, preeclampsia with severe features, superimposed preeclampsia and eclampsia) and cardiovascular mortality, including deaths related to heart disease and stroke, from the time of delivery up until the first calendar year postpartum. Using a large, nationally representative sample of patients in the US (Nationwide Readmission Database [NRD]) with follow-up in the first year postpartum, they were able to address the limitations of prior work.3 Importantly, by highlighting the associations between specific subtypes of HDP (except gestational hypertension) and cardiovascular mortality (adjusted hazard ratios range: 1.96 to 58.55, with the highest risk for eclampsia), their work illustrates the persistent contribution of HDP to maternal mortality during pregnancy and postpartum despite advances in obstetric care.3 While this study emphasises the short-term cardiovascular burden associated with subtypes of HDP, the observed associations between HDP and cardiovascular mortality may have been underestimated. The restriction of the study population to women without pre-existing cardiovascular disease (of note, patients with congenital heart disease were not excluded) could have attenuated the observed associations since prior evidence has shown an increased risk of mortality among women with structural heart disease and co-occurring HDP.4 Moreover, while cardiovascular diagnoses included in the outcome were comprehensive, they did not capture indicators of ruptured aneurysm.5 Lastly, the sensitivity of ICD-9 and 10 codes for cardiovascular severe maternal morbidity as a composite outcome, particularly in the context of preeclampsia, has previously been found to be low, ranging from 10% to 50%.5 Therefore, cardiovascular events may have been misclassified, potentially attenuating the true association between HDP and cardiovascular-related deaths. To assess the cardiovascular nature of reported deaths, all pregnancy-related in-hospital deaths in the same calendar year as the birth were identified and stratified into those with co-incident heart disease and stroke. However, the direct link between cardiovascular events and mortality could not be ascertained. While individuals with cardiovascular diagnoses captured prior to pregnancy were excluded from the study population, cardiovascular diagnoses captured during pregnancy included in the “heart disease-related deaths” category could include diagnoses corresponding to prevalent, chronic conditions (e.g. ischaemic heart disease and atherosclerotic heart disease), which did not necessarily represent acute events. Although deaths in the context of these cardiovascular conditions were being captured, these deaths may not have directly resulted from them. For instance, upon review of granular clinical data during a confidential inquiry, a patient with stable hypertensive heart disease who died following severe sepsis may have been classified as having had an infection-related death. While her death occurred in the context of an underlying heart condition, she may not have been considered as having had a heart disease-related death. Since centralised reporting systems and confidential enquiries are not systematically in place to perform a detailed assessment of pregnancy-related deaths, we may need to continue to rely on administrative datasets (like the NRD) to assess causes of maternal mortality at the population level with a certain degree of misclassification. While the authors highlighted the increased accuracy of using in-hospital mortality based on administrative datasets such as the NRD compared to estimates based on vital statistics data using the standardised pregnancy checkbox on death certificates,3 the accuracy of underlying causes of maternal mortality identified using these datasets must be optimised prior to their use for surveillance purposes. Severe maternal morbidity is comprised of a set of unexpected maternal outcomes related to pregnancy, labour, childbirth and the postpartum period resulting in severe illness, prolonged hospitalisation and/or long-term disability.6 Since severe maternal morbidity may be on the pathway to maternal mortality,6 the use of indicators of severe cardiovascular morbidity (e.g., acute myocardial infarction), in addition to underlying chronic cardiovascular diagnoses (e.g., coronary artery disease), could improve the accuracy of reporting of causes of mortality, emulating the cause-of-death section of death certificates.7 However, this does require further exploration using well-designed validation studies. While indicators of severe maternal morbidity have been standardised for use in surveillance and research, indicators used to capture causes of maternal mortality are lagging. As a result, indicators used to identify cardiovascular mortality cases differed from those used to identify cardiovascular subtypes of severe maternal morbidity events, as defined by the Centers for Disease Control and Prevention.5, 8 To ensure accurate reporting and implementation of preventative measures geared towards reducing cardiovascular morbidity and mortality, there is an urgent need to harmonise the assessment of cardiovascular causes of severe maternal morbidity and mortality. Finally, due to the process for data collection in the NRD, the ability to observe events up to 1 year postpartum was not possible for all women delivering after January, resulting in missed events. This raises concerns regarding the utility of data sources that capture events based on calendar year, since the postpartum period is known to be a high-risk period for mortality for individuals with HDP and cardiovascular complications.9 However, the increased risk in the earlier periods of follow-up (<60 days postpartum) suggests that the extent of the attenuation of the observed associations may be less of a concern. While, Lee and colleagues3 demonstrated the added value of conducting maternal mortality surveillance using routine administrative datasets with extended follow-up beyond 42 days postpartum, the results need to be interpreted in light of the potential limitations arising from the quality of the data source. The strong association between HDP subtypes and cardiovascular mortality in this study highlights the need to extend postpartum care and health insurance coverage for pregnant individuals later than the early postpartum period to prevent maternal cardiovascular deaths.3 In addition, concerted efforts are required to reduce the risk of cardiovascular-related morbidity and mortality in women with HDP, including strategies to raise awareness among clinicians and patients regarding the heightened risk in these individuals, interventions to harmonise cardiovascular care across the pregnancy continuum, and increased availability of programs that allow for continual follow-up in the first year postpartum to align with current clinical recommendations.10 The findings by Lee and colleagues3 expand the current evidence on the association between HDP subtypes and cardiovascular mortality, including the postpartum period, a crucial period for the prevention of HDP-related short and long-term cardiovascular morbidity and mortality. This work highlights the need for future research focused on the validation of underlying causes of maternal mortality and the need for higher quality sources of epidemiological data to optimise surveillance of maternal mortality. IM and SMG both contributed to drafting and critically rivising this manuscript. Isabelle Malhamé is a Clinician Scientist specialised in Obstetric Medicine and an Assistant Professor in the Department of Medicine at McGill University. Her research program focuses on the reduction of severe maternal morbidity, with a particular interest for cardiovascular complications. Her work is articulated around the following research themes across the pregnancy continuum: risk prediction, innovation in the diagnosis and management of high-risk conditions, and quality improvement and patient safety. Sonia M. Grandi is a Scientist at the Hospital for Sick Children and an Assistant Professor in the Division of Epidemiology at the Dalla Lana School of Public Health at the University of Toronto. As a perinatal epidemiologist and methodologist, her research focuses on the contribution of preconception and perinatal factors to the short- and long-term cardiometabolic health of mothers. As part of her work, Dr Grandi has examined the role of adverse pregnancy outcomes in long-term cardiovascular health in mothers and the development of effective screening tools and targeted strategies for the early identification of women at risk of future cardiovascular disease. She serves as a junior editor of Paediatric and Perinatal Epidemiology. Not applicable. None declared. Not applicable.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.