Using data from the UK Biobank, this study systematically investigated the relationship between stillbirth and diabetes, CVD, all-cause mortality, as well as CVD mortality. Results indicated that women with a history of stillbirth had a significantly higher risk for most outcomes compared to those without. The association between stillbirth and diabetes, CVD, all-cause, and CVD mortality remained significant regardless of spontaneous abortion or pregnancy termination history. Notably, for participants with annual incomes above 31,000 £, stillbirth was not significantly linked to all-cause or CVD mortality.
This study found that women who experienced stillbirth had an increased risk of diabetes and CVD, including its various subtypes. Although a previous studies found that pregnancy loss (including miscarriage and stillbirth) was associated with a higher risk of CVD [14, 15], unfortunately, there were too few stillbirths to allow for corresponding analyses for stillbirths. Similarly, most previous studies have included pregnancy loss as a total exposure factor, which includes not only stillbirths but also spontaneous and induced abortions [4,5,6,7]. In addition, a limited number of studies have reported on the risk of developing diabetes following pregnancy loss. For example, a study from China indicated that women with a history of stillbirth had an increased risk of diabetes [16]. Consist findings were reported in a study involving a western population, where having more than two miscarriages was associated with an approximately two-fold higher risk of diabetes, however, this study did not provide evidence specific to stillbirth [17]. In addition, most of the studies relied on national registry data and were limited in scope regarding behavioral covariates and socioeconomic modifiers. Our study adds to this evidence base by using prospective cohort data with detailed lifestyle and socioeconomic variables, allowing for refined analyses of potential effect modifiers. Moreover, by evaluating multiple cardiometabolic and mortality outcomes simultaneously, we provide a broader understanding of the post-stillbirth health trajectory. Our analysis also highlights the modifying role of income and lifestyle factors, offering insights into potential intervention points to reduce health disparities among women affected by stillbirth.
Stillbirth, diabetes, CVD, and CVD mortality may involve common underlying mechanisms, such as disorders of glucose metabolism [18, 19]. In addition, stillbirth may influence the development of CVD by bypassing other pathways of diabetes. For example, antiangiogenic status may increase the risk of stillbirth [20], and antiangiogenic proteins may be a new pathogenesis of CVD in the absence of diabetes [21]. In exploring the relationship between stillbirth, diabetes, and CVD, the deficiency of nerve growth factor (NGF) plays a crucial role. Studies indicate that women who experience stillbirth may face a decrease in NGF levels, which can lead to the apoptosis of pancreatic β-cells, thereby affecting insulin secretion and glucose metabolism, ultimately increasing the risk of diabetes [22, 23]. Additionally, in diabetic patients, the reduction of NGF levels is closely associated with the apoptosis of pancreatic beta-cell, a process that results in impaired insulin secretion through the inhibition of the PI3K and AKT signaling pathways, activation of c-Jun kinase, and reduction of Bcl-Xl protein [24]. On the other hand, the decrease of NGF can also activate the C-Jun N-terminal kinase pathway, induce the expression of apoptosis-related genes, such as promoting the release of cytochrome c, and then activate the apoptotic cascade. In addition, decreased levels of NGF were accompanied by decreased expression of the anti-apoptotic protein Bcl-xL, further weakening cell survival [25]. These molecular changes eventually lead to impaired insulin synthesis and secretion function, a decrease in the number of islet beta cells, and the destruction of the islet structure, which triggers or exacerbates the development of diabetes. In terms of CVD, patients with ischemic heart disease exhibit significantly lower NGF levels compared to healthy individuals, which may correlate with myocardial cell necrosis and HF [26, 27]. Research has shown that NGF can improve cardiac function, reduce myocardial apoptosis and fibrosis, and protect the heart by promoting autophagic flux and attenuating protein ubiquitination [23]. In summary, the deficiency of NGF establishes a complex mechanistic pathway in the occurrence and development of stillbirth, diabetes, and CVD, warranting further in-depth investigation.
Emotional and lifestyle changes may also be an important factor in the increased risk of diabetes after stillbirth. Stillbirth, especially when experienced multiple times, tends to have a negative psychological impact on women, increasing anxiety and mental stress, which in turn affects behavior and health habits [17]. Studies have shown that women who experience stillbirths are more likely to be overweight than those who experience live births [28], which may be due to psychological stress that alters eating habits. Also, about 10% of women who experience stillbirth develop acute stress disorder [29]. Repeated stress exposure may trigger chronic stress, leading to disturbances in glucose metabolism, neuroendocrine disorders, and a prolonged low-grade inflammatory response. Depression, persistent stress and early adverse experiences are strongly associated with the development of diabetes [30]. Similarly, obesity and depression can also lead to an increased risk of CVD [31]. Women with a history of stillbirth may benefit from targeted screening and prevention strategies. Higher income may reduce the risk of cardiovascular death and all-cause death associated with stillbirth through a variety of mechanisms, including access to health care, stress management, and healthy behaviors. Higher-income women often have access to better medical resources, which allows them to receive early screening, preventive interventions, and better health management, such as diabetes and hypertension [32]. High-income groups are often able to enjoy regular health checkups, medication and preventive care, which reduces the long-term risk of serious health problems such as diabetes and CVD. In addition, higher income individuals generally experience lower levels of psychological stress, which has a positive impact on cardiovascular health [33]. Stress is a known risk factor for heart disease, and chronic stress can lead to activation of the hypothalamic–pituitary–adrenal axis, promoting inflammation and endothelial dysfunction, both of which are associated with the development of CVD [34, 35]. Conversely, low-income women may face greater stress due to financial hardship, lack of social support, or limited medical resources, which may exacerbate the risk of cardiovascular death and all-cause mortality [36]. Higher-income women may also be more likely to adopt healthy lifestyles, such as regular exercise, a balanced diet and lower rates of smoking, all behaviors that help reduce the risk of CVD and early death [37].
Our findings highlight significant heterogeneity in CVD and all-cause mortality outcomes related to stillbirth across different income levels. Specifically, among participants with annual incomes above 31,000 £, stillbirth was not significantly associated with all-cause or CVD mortality, suggesting that higher income may mitigate some health risks linked to stillbirth. No significant association between stillbirth and all-cause and cardiovascular death was observed in people without hypertension, possibly because hypertension plays an important mediating or modifying role between stillbirth and subsequent health outcomes. Hypertension is closely related to vascular endothelial dysfunction, inflammation and other pathological mechanisms, which may also be involved in the occurrence of stillbirth and affect long-term health risks [38]. Therefore, in individuals without hypertension, the absence of this pathological basis may have weakened the effect of stillbirth on the risk of death, and thus did not show a statistically significant association.
This aligns with a recent global burden of disease study indicating that stillbirth rates tend to decline as socioeconomic development increases. However, the highest burden of stillbirth remains concentrated in sub-Saharan Africa and South Asia, where socioeconomic challenges persist. Furthermore, studies have shown that the majority of stillbirths in these high-burden regions occur in rural areas with low Healthcare Access and Quality indices [39, 40]. These findings suggest that low-income families or regions may bear a disproportionate burden of stillbirth-related mortality, underscoring the need for targeted interventions, which provides a scientific basis for governments and public health organizations to develop strategies and allocate resources effectively to reduce the burden of stillbirth and its associated health outcomes. Emerging evidence indicates that a proportion of unexplained stillbirths may be linked to fetal cardiac channelopathies, such as long-QT and short-QT syndromes. These conditions, often caused by mutations in genes like KCNQ1, KCNH2, and SCN5A, can result in fatal fetal arrhythmias even in the absence of structural cardiac abnormalities or fetal growth restriction. Additionally, maternal long-QT syndrome may impair placental or uterine function, thereby increasing the risk of fetal loss [41].
This work has two main strengths as follows: UK Biobank is a large and long-term follow-up cohort with large sample and abundant data which enhances the statistical validity of study. In addition, we explored the association of stillbirths with multiple outcomes and enriched the study of the experience of stillbirth on women’s long-term health outcomes. However, this study also has some limitations. First, because of the small sample size of CVD mortality, the results may be affected by the complexity of the model, which may reduce its stability. Second, although the robustness of the results was enhanced by sensitivity analyses that excluded outcomes occurring within two years, causality could not be established due to the observational design of the study. The present study revealed an association between stillbirth and these outcomes, but the underlying mechanisms have not been clarified. Third, the majority of the study population was White (95.1%), which limits the extrapolation of the results to other racial groups. Differences based on race and ethnicity may lead to different health outcomes, and this area needs further research. Fourth, the study was not able to obtain specific information on the occurrence of stillbirth. Because most of stillbirths occurred many years ago, changes in an individual’s health status, for example BMI measurement or lifestyle estimates at recruitment rather than the onset of stillbirths, and environment may affect the association between stillbirths and health outcomes, limiting the interpretation of temporal relationships. Fifth, because it is difficult to confirm whether participants had gestational diabetes at each pregnancy, some residual confounding may persist. However, we further excluded participants with pre-enrollment diabetes to minimize this potential bias. Sixth, likewise, specific information of medical services during pregnancy and frequency of medical visits is unavailable. In order to minimize the impact of these factors, we further added TDI and income level covariates to evaluate associations of stillbirths with target outcomes and obtained consistent results. Seventh, the number of stillbirths occurring among the participants in this study was concentrated at a low frequency, with a smaller sample experiencing multiple stillbirths, which limited the precision of our stratified analysis and quantitative assessment of the association between the number of stillbirths and health outcomes. Eighth, stillbirth history was obtained through self-report, which may be subject to recall bias, particularly for events that occurred many years prior to baseline. This introduces the potential for both underreporting and misclassification of exposure. If the accuracy of recall is associated with the participant’s health status, differential misclassification could occur and may bias the observed associations in either direction. However, in the absence of such systematic differences, any misclassification is likely to be non-differential and may have attenuated the true associations. Ninth, we acknowledge that certain important obstetrical and reproductive health variables, such as gestational age at the time of stillbirth, history of recurrent pregnancy loss, use of assisted reproductive technologies, and specific pregnancy complications (including preeclampsia and gestational diabetes), were not available in current dataset. These unmeasured factors are biologically linked to both stillbirth and long-term health outcomes and may have introduced residual confounding. Therefore, their absence should be considered when interpreting the findings of this study.
Given the significant associations found in this study between stillbirth and the risk of diabetes, CVD, and death, more aggressive health management for women with a history of stillbirth is recommended. This group of people should be included in the high-risk group after childbirth for systematic metabolic and cardiovascular health assessment. Diabetes screening recommends a fasting blood glucose or glucose tolerance test every 1 to 3 years, depending on the individual’s weight, family history and other risk factors. For CVD, blood pressure, lipids, and lifestyle factors should be assessed annually, combined with an electrocardiogram or echocardiogram if necessary. In addition to medical screening, psychological intervention and lifestyle management, including weight loss, nutritional guidance and exercise intervention, should be strengthened to reduce the long-term risk of chronic diseases.