Postpartum Hemorrhage Patients: Practical Research For Future Cardiovascular Events Risk
Overview
The study aimed to explore the link between postpartum hemorrhage (PPH) and future cardiovascular disease. This population-based retrospective cohort study utilized record linkage between the Aberdeen Maternity and Neonatal Databank (AMND) and Scottish healthcare datasets. The research was conducted in the Grampian region of Scotland and included a cohort of 70,904 women who gave birth after 24 weeks of gestation from 1986 to 2016. Extended Cox regression models were employed to examine the association between experiencing one or more instances of PPH in any birth (whether first or subsequent) and the onset of cardiovascular disease, while adjusting for sociodemographic, medical, and pregnancy-related factors. The primary outcome measured was cardiovascular disease, identified through the prescription of specific cardiovascular medications, hospital discharge records, or death attributable to cardiovascular disease.
Introduction
The incidence of postpartum hemorrhage (PPH) has been rising in many high-income countries (HICs) over the past few decades. Globally, PPH accounts for approximately one-quarter of all maternal deaths, translating to around 70,000 maternal deaths each year. Although maternal mortality due to PPH is rare in HICs, the increasing incidence of PPH continues to pose a significant risk for maternal morbidity. The well-established associations between PPH and acute physical complications, such as hypovolemic shock, sepsis, and disseminated intravascular coagulation, highlight the immediate dangers following severe hemorrhage. However, the long-term consequences of postpartum hemorrhage, particularly those extending beyond five years post-event, have been less frequently studied.
Recent systematic reviews suggest that women who experience PPH may face long-lasting adverse health outcomes, including cardiovascular disease (CVD). Two studies within these reviews identified a link between postpartum hemorrhage requiring blood transfusion and subsequent hospitalization for CVD. However, they did not find a similar association for PPH cases that did not necessitate transfusion. Conversely, a comparable English cohort study, utilizing linked primary care databases, examined the long-term risk of hypertension and CVD. This study found no significant difference in CVD risks over a median follow-up period of four years. Notably, the literature lacks comprehensive studies that consider the potential variations in the PPH-CVD association over time.
Biologically, a potential connection between postpartum hemorrhage and future CVD is plausible. Extreme blood loss during childbirth can cause significant hemodynamic instability, potentially leading to hemorrhagic shock and end-organ damage. This condition may impair cardiac function, raising the likelihood of developing CVD, particularly ischemic heart disease (IHD), in later life. Given the considerable health burden of CVD, which affects approximately 7.6 million people in the UK alone, the implications of postpartum hemorrhage on long-term cardiovascular health warrant further investigation.
To address these gaps, we conducted a population-based, record-linked cohort study to explore the relationship between primary PPH and the subsequent development of CVD in a Scottish population. Our study aimed to estimate the incidence and relative risk of CVD among women who experienced at least one episode of PPH compared to those who never had PPH. Additionally, we sought to determine whether the association between PPH and CVD is modified by the mode of delivery or the presence of hypertensive disorders of pregnancy (HDPs).
Also read; Postpartum Hemorrhage To Reveal Increased Risk For Postpartum Depression
By leveraging data from the Aberdeen Maternity and Neonatal Databank (AMND) and linked Scottish healthcare datasets, we identified a cohort of 70,904 women who gave birth after 24 weeks of gestation between 1986 and 2016. Using extended Cox regression models, we examined the association between PPH and subsequent CVD, adjusting for sociodemographic, medical, and pregnancy-related factors. Our primary outcome measures included the incidence of CVD, identified through the prescription of selected cardiovascular medications, hospital discharge records, or death attributable to CVD.
The investigation of PPH’s long-term health effects, particularly its potential impact on cardiovascular health, is crucial for developing comprehensive maternal care strategies. Our study aims to provide valuable insights into the lasting health consequences of PPH and inform clinical practices to better support women’s long-term health outcomes following severe postpartum hemorrhage.
Methods
Inclusion Criteria
– Population: All women who gave birth after 24 completed weeks of gestation at Aberdeen Maternity Hospital from January 1, 1986, to December 31, 2016.
– Births: Included both singleton and multiple births.
– Healthcare Registration: Only permanent registered patients with NHS Scotland were included to minimize the loss to follow-up.
Exclusion Criteria
– First Birth Not Recorded: Women whose first birth was not recorded in the Aberdeen Maternity and Neonatal Databank (AMND) were excluded to ensure accurate data linkage and exposure classification.
– Conflicting Parity Information: Women with conflicting parity information (e.g., the same parity recorded for two birth records) were excluded.
– Births Outside Grampian Region: Women who had at least one of their births outside the Grampian region (e.g., non-consecutive parity records) were excluded to maintain data consistency.
– Pre-existing Cardiovascular Disease (CVD): Women who had a CVD event before their first birth or before their first postpartum hemorrhage were excluded to ensure the temporality of outcome measurement.
– Post-CVD Births: Birth records occurring after a CVD outcome event were excluded to focus on the impact of postpartum hemorrhage on subsequent CVD risk.
Analysis
Maternal sociodemographic and clinical characteristics were analyzed based on postpartum hemorrhage (PPH) status, utilizing frequencies and percentages for stratification. To examine the association between postpartum hemorrhage and CVD-free survival, univariable Cox proportional hazards regression was employed.
To further investigate this association, multivariable Cox proportional hazards regression models were constructed. These models, which estimated adjusted hazard ratios (aHRs), were adjusted in a stepwise manner for potential confounders identified through a causal diagram. Initially, all models were adjusted for the year of birth to account for temporal trends. Following this, adjustments were made for:
- Maternal sociodemographic factors (Model 1)
- Maternal medical history, including self-reported CVD history (Model 2)
- Pregnancy-related factors (Model 3)
Potential confounding variables were included based on their biological plausibility, association with postpartum hemorrhage, and impact on CVD risk in the unexposed group, provided they were not on the causal pathway between postpartum hemorrhage and CVD. Data from each birth for individual women were used to adjust for confounding in the multivariable models.
Log-log plots assessed the consistency of the hazard ratio for the association between PPH/severe PPH and CVD over the follow-up period. The global proportional hazards test evaluated the proportional hazards assumption for exposure variables and confounders. Extended multivariable Cox regression analyses were conducted to address any violations of the proportional hazards assumption, estimating aHRs for various follow-up periods post-birth. Potential effect modification by hypertensive disorders of pregnancy (HDPs) and mode of birth was also examined.
Two sensitivity analyses using Model 3 explored the PPH-CVD association further:
- Excluding women with a self-reported history of CVD before their first birth.
- Restricting the outcome measure to CVD-related hospitalizations.
For covariates with missing data, a ‘missing’ category was created. Given that missing data was less than 5% and unlikely to be associated with the CVD outcome, a complete case analysis was performed. Analyses were conducted at the individual woman level, incorporating all births to enhance the study’s generalizability. Stata 16 (StataCorp LLC, College Station, TX, USA) was used for all data analyses.
Results
Data Overview
The linked dataset included information on 125,022 birth records from 72,593 women who gave birth between 1986 and 2016. After excluding certain records, the final cohort comprised 121,731 births from 70,904 women. Among these, 25,177 women (36%) experienced at least one primary PPH, with 4,799 women (6%) experiencing severe postpartum hemorrhage. Women with postpartum hemorrhage were more likely to be aged 30 or older at their first birth, belong to Black, Asian, or other ethnic backgrounds, and have a higher BMI at their first antenatal visit. Within the postpartum hemorrhage group, 20,416 (81%) had one PPH, 4,320 (17%) had PPH twice, and 441 (2%) had PPH three or more times.
Outcome Events
Prescription records accounted for approximately 91% of CVD outcome events, CVD-related hospitalizations for 9%, and CVD-related deaths for less than 1%. Around 13% of women with prescriptions as their first CVD event subsequently had a CVD-related hospitalization. Over half of the women with CVD-related hospitalizations as their first event were later prescribed CVD medication. Both groups observed small numbers of CVD-related deaths.
Univariable and Multivariable Cox Proportional Hazards Models
Univariable analysis revealed that women with at least one postpartum hemorrhage were 6% more likely to develop a CVD outcome during the follow-up period (HR 1.06, 95% CI 1.02–1.11). After adjusting for year of childbirth (base model), women with at least one postpartum hemorrhage were 29% more likely to develop a CVD outcome (aHR 1.29, 95% CI 1.23–1.35). Further adjustments for maternal sociodemographic factors (model 1) and medical history (model 2) did not significantly alter this association. Adjusting for pregnancy-related factors (model 3) attenuated the association but it remained statistically significant (aHR 1.12, 95% CI 1.06–1.18). Similar results were found for severe postpartum hemorrhage.
Extended Cox Regression Models for Different Time Periods
In the first year after birth, the fully adjusted model (model 3) showed that women with at least one PPH were almost twice as likely to experience a CVD event compared to women without PPH (aHR 1.96, 95% CI 1.51–2.53). Beyond the first year, the association remained but weakened over time (aHR 1.19 at 2–5 years; aHR 1.17 at 6–15 years). Beyond 15 years, women with at least one PPH were 36% less likely to have a CVD event compared to those without PPH (aHR 0.64, 95% CI 0.55–0.75). The association between severe PPH and CVD over time followed a similar pattern.
Effect Modification
There was strong evidence of effect modification by HDPs in the fully adjusted model, with higher aHRs for the association between PPH/severe PPH and CVD among women who had at least one birth affected by HDPs (p = 0.003). There was weak evidence of effect modification by mode of birth (p = 0.09).
Conclusion
Compared to women who have never experienced a postpartum hemorrhage, those with a history of at least one PPH, regardless of its severity, have double the risk of developing cardiovascular disease (CVD) within the first year postpartum. This elevated risk continues, albeit reduced, for up to 15 years after childbirth, even after accounting for potential confounding factors. The risk may be marginally higher among women with hypertensive disorders of pregnancy (HDPs). PPH should be recognized as an early indicator of future CVD risk, warranting further research into the long-term health implications for women following postpartum hemorrhage.
Oncology Related Tools
Other
Latest Research
