Prenatal care for patients with cardiac disease is complex, often involving close surveillance, intensive monitoring, and peripartum interventions that can result in mother infant separation after birth [1,2]. Though these experiences can be distressing for patients, little is known about postpartum depression rates in this high-risk population. Characterizing postpartum depression and identifying risk factors might inform risk reduction strategies. Therefore, we sought to conduct an exploratory study to evaluate clinical factors associated with a positive postpartum depression screen in pregnant people with cardiac disease.

This was a retrospective cohort of people with congenital and/or acquired cardiac disease during pregnancy who delivered at a single quaternary institution between 2014 and 2020. People with cardiac disease who had Edinburgh Postpartum Depression Scores (EPDS) documented at the postpartum visit – which typically occurs 4-6 weeks after delivery – were included. If patients had more than one pregnancy, only the first pregnancy at our institution with an EPDS score was included. The primary outcome was a positive depression screen, defined as EPDS score ≥10 [3]. Potential exposures associated with a positive postpartum depression screen were selected a priori based on prior literature and included: composites of adverse maternal cardiac and neonatal events; measures of cardiac disease severity, such as World Health Organization (WHO) Classification at the beginning of pregnancy or anticoagulation with heparin or enoxaparin; substance use; delivery complications such as blood transfusion or postpartum readmission; breastfeeding status at hospital discharge; and mother-infant separation during the delivery hospitalization due to level of care (e.g. maternal or neonatal ICU admission).

Secondary outcomes were responses to a follow-up survey that was sent to all English speaking patients who provided contact information and consented to future research. This survey included questions about cardiac status, mental health, and infant development since the index delivery. Wilcoxon rank-sum and Fisher’s exact tests were used to compare clinical factors between patients with and without positive postpartum depression screens.

Of 126 eligible cardiac patients, 23 (18.3%) had a positive postpartum depression screen. Patients with a positive postpartum depression screen were more likely to have had antepartum anticoagulation with heparin or enoxaparin (56.5% versus 26.2%, p=0.007), blood transfusion (8.7% versus 0%, p=0.032), and maternal-infant postpartum separation (52.2% versus 28.2%, p=0.047, Table 1). There were trends toward higher-risk features in the positive postpartum depression screen group based on higher WHO scores, more antepartum hospitalizations, and greater preterm birth and preeclampsia rates, though these were not significantly different. Although there was a high prevalence of pre-pregnancy depression or anxiety in the whole cohort (N=26, 20.6%), there was no significant difference in positive postpartum depression screens for these patients (17.4% versus 21.4%, p=0.78).

Of the 126 patients included in this study, 71 were eligible for the follow-up survey and 29 responded (41% response rate). Among patients with a positive postpartum depression screen who responded to the survey, the median time from delivery was 1.8 years (Table 2). On the survey, 50% of patients with a positive postpartum depression screen reported being formally diagnosed with anxiety or depression and 33.3% reported child development problems such as speech delays.

18.3% of patients in this cohort with cardiac disease in pregnancy had a positive postpartum depression screen. For context, the prevalence of depression in the general postpartum population has been estimated between 6-13% [4,5], while the lifetime prevalence of depression in non-pregnant adults with cardiac disease is closer to 30% [6]. Interestingly, 20.6% of all people in this cohort had a pre-pregnancy diagnosis of depression or anxiety but they were not more likely to have had a positive postpartum depression screen. The majority of patients with a positive postpartum depression screen therefore did not have a prior psychiatric diagnosis; taken together, this means that 45 (35.7%) of all pregnant people with cardiac disease in this cohort had either pre-pregnancy depression or anxiety or a positive postpartum depression screen. This highlights the significance of future research to understand risk factors for depression in this population, as identification of actionable risk factors may help reduce the mental health burden these patients are facing.

A positive postpartum depression screen was significantly associated with antepartum anticoagulation with heparin or enoxaparin, blood transfusion, and maternal-infant separation during the delivery admission. The blood transfusion association should be interpreted with caution given small frequencies. Our results suggest that high risk delivery features, particularly those that result in maternal-infant separation in the immediate postpartum period, set the stage for patients with cardiac disease to develop postpartum depression. It is also notable that half of the people with a positive postpartum depression screen in this study who responded to a follow up survey reported ultimately receiving a formal diagnosis of anxiety or depression after delivery.

Strengths of this study include granular patient-level data and breadth of clinical exposures potentially associated with postpartum depression. Despite these strengths, our results must be interpreted in the context of our study design. Although this cohort represents 7 years of consecutive data from a single institution, our small sample size precludes additional analyses to adjust for potential confounders. Furthermore, findings from a single academic institution may not be generalizable to other populations. That being said, the results from our study can be used as pilot data for future research among larger, more generalizable cohorts to further examine the associations we identified.

Our results highlight the importance of screening for postpartum depression in patients with cardiac disease, especially those requiring antepartum anticoagulation or maternal-infant separation. While our study is hypothesis generating, these clinical factors warrant further investigation in larger cohorts. Addressing maternal-infant separation in particular may require operational planning, particularly for large hospital systems with separate obstetric, pediatric, and/or adult facilities.

Due to its retrospective nature, the Stanford Institutional Review Board (IRB) waived the need for informed consent. For this type of study, formal consent is not required.

This study was approved by the Stanford University Institutional Review Board (IRB) prior to its initiation. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the Helsinki declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.

The authors declare that they have no conflict of interest.

Dr. Panelli’s time is partially funded by an NIH K12HD103084.